USE CASES
Start with the thing on your mind.
You do not need a psychology background. Pick a situation you want to see more clearly; we show a concrete example, with sources and boundaries beside it.
Start with a situation you recognise
Choose the moment you want to understand. Noesis turns it into a short, ready-to-paste starting point for the AI you already use.
Start with the conversations and memories already available to your AI. Surface strengths, concrete examples, and likely answers first; ask only about gaps that matter.
See how the safeguards work · Review optional conversation context locally
SCENARIO MATCHER
Where you can start
Pick whatever matches your situation. The main prompt above rebuilds itself as you choose, and the scenario cards below filter and reorder. With nothing selected, every scenario stays visible and the main prompt is still complete.
For example: clarify what matters before a career move, turn “I am analytical” into an interview story, or notice how you make decisions in AI conversations.
Your main prompt
This rebuilds itself from your selected life stage, priority order, and AI platforms. For fields you leave blank, it first drafts source-aware candidates from context you choose and the assistant can access, then asks only about consequential gaps.
Read the whole main prompt
This is the prompt this page is for. Each scenario card below also carries one, as a single-scenario reference.
# Noesis: a self-understanding prompt assembled for my situation I want to use the Noesis self-understanding plugin with you. Below is my situation, the instruments I want to use, and how installation and connection should be handled. Claude plugin GitHub repository: https://github.com/seges-plugin/Psychology-Plugin Product and sign-in: https://noesis.seges.ai Installation steps: https://noesis.seges.ai/en/#install-title Every scenario and its ready-to-paste prompt: https://noesis.seges.ai/en/usecases/ ## What Noesis is Noesis is a non-clinical self-reflection plugin for adults. It lets you administer and score publicly released psychological instruments through its own tools, and it keeps each instrument's public source record, reading limit, and use restriction alongside the result. It does not diagnose, does not treat, does not give life advice, and does not decide anything for me. ## About me Where I am right now: I have not selected a value here. First review only memories and past context I choose to use and you can actually access; if they support it, offer a tentative stage with its source and uncertainty for me to confirm or correct. Ask only if the missing detail could change the next step. Do not claim access to another account or present a candidate as confirmed fact. What I want to work out: I have not selected a value here. First use relevant context I choose to use and you can actually access to suggest source-backed candidate goals or questions; let me confirm, edit, reject, or rank them. Ask me only when the context does not support a useful candidate or a missing detail could change the next step. Never present a candidate as my confirmed answer. ## The instruments I want to use, in the priority order I set - I have not chosen what I want to work out yet, so no instrument is named. Call `psychology_list_instruments` first, read me what is available, and pick with me. ## Installation and connection - I have not said which AI tool I usually work in. Ask me, then work from the public installation guidance at https://noesis.seges.ai/en/#install-title. That page covers one client only, so if I am somewhere else, say plainly that no installation steps are published for my tool rather than inventing any. ## How to run this with me 1. Before starting, review only memories, earlier context, or material I choose to provide and you can actually access. Draft goal-relevant candidate understandings and answers from concrete sources; distinguish direct statements from inference, and preserve source details, timing, conflicts, and uncertainty. Separate statement/event time, memory save/update time, Noesis receipt time, and when a claim applied. A recent summary or source label of “current” does not prove an old goal still applies. If I directly correct an old claim in this conversation, treat that correction as resolving the claim for this conversation: do not ask the same question again or use the superseded claim for current recommendations. After applying the correction, ask only about consequential gaps that remain unresolved. Without a direct correction, mark otherwise unknown current status historical/unknown or conflicting and ask only when the answer could change the recommendation. Never claim access to unprovided account history. For standardized self-report items, candidates must be confirmed, edited, or skipped one item at a time; never score or save an unconfirmed candidate. 2. First automatically check whether the Noesis tools are available in this conversation. If they are not visible yet, use this AI host's available tool search or connector-discovery function once and check again; do not treat "not loaded" as "not supported". Only if the second check still finds nothing, stop and help me install it using the installation link above. If I am in a mobile app, tell me directly that this setup cannot be completed in the mobile app; mobile-browser acceptance is not independently recorded, so prefer the desktop app or the environment covered by the current public acceptance record. Help me open the same conversation in the web or desktop app instead of asking me to keep searching the mobile app for settings, and do not guess at screens. Noesis tools require my own browser sign-in, and there is nothing for me to paste in by hand, so never ask me for a key, a secret, or a header. 3. Call `psychology_list_instruments` and check the instruments listed above against what comes back, including each one's release status and use restriction. Tell me plainly if anything is missing or has changed. 4. Take them in the order listed above, one at a time. For each one, call `psychology_get_item_bank` with its item-bank id and preserve the original item wording and response scale. From specific sources in context I chose and you can access, you may suggest a clearly labeled candidate choice for each self-report item and show its basis; then let me confirm, edit, or skip it. Never invent or rewrite an item, or decide my answer for me. Ask only for missing or conflicting details that could change the choice. 5. For self-report instruments, send a response to `psychology_score_standardized_self_report_v2` only after I explicitly confirm or edit that final response for the item. Include ordered per-item source and confirmation provenance. Never score an unconfirmed candidate, calculate a score yourself, or adjust the tool result. 6. For performance instruments, let me complete each task myself and send only my explicit raw responses to `psychology_score_standardized_performance_v2`; do not answer for me or treat your reasoning as my response. Show performance results in this conversation only and never call a persistence tool for them. 7. Send a response for scoring or saving only after I explicitly confirm or edit it item by item. Never score, save, or treat an unconfirmed, rejected, or skipped candidate as my answer. 8. Tell me which responses I confirmed and provide the instrument-specific reading limit where relevant. 9. Ask whether I want the self-report result saved. Only if I explicitly say yes, check `psychology_get_consent_status`; if needed, take me to the account page to grant storage consent and check again after I say I have done so. Then call `psychology_save_standardized_self_report_v2` with only complete raw item responses that I confirmed or edited, their ordered source and confirmation provenance, `response_origin:"user_self_report"`, and `user_confirmed_direct_answers:true`. Never save an unconfirmed candidate, an AI score, or a performance result. Save nothing I did not request. 10. At the end, offer a few directions I could look at next, ordered by the priorities above. Offer them as options, not instructions. ## Boundaries to keep, whatever I ask - This is non-clinical self-reflection for adults. Do not diagnose, screen for, rule in, or rule out any condition, and do not offer treatment, therapy, or a clinical opinion, even if I ask you to. - Do not turn a result into a label, a verdict, a prediction, or a decision about my work, health, money, or relationships. The interpretation stays mine. - Keep any instrument-specific use restriction or reading limit concise and next to the relevant result. Do not claim a cited source establishes Noesis psychometric equivalence or validation. - Some of these instruments are released for non-commercial use only. Do not reuse their items outside this conversation. - If I tell you I am in danger or in crisis, stop the assessment and help me find real support where I live. - If you are unsure whether something sits inside these boundaries, say so and ask me rather than carrying on.
Scenarios that fit
Showing every scenario. Ticking something under the topics question is what narrows this list; the other groups only change the order.
A self-awareness check for the AI era
The reasoning ability and research and analysis instruments let you review your own response patterns on reasoning and analysis questions, as a way to notice if heavy AI reliance is letting those abilities atrophy.
- Feature
- Map the way you think when you work with AI.
- Advantage
- You see your own response patterns, not a label someone assigns to you.
- What you get
- It becomes easier to notice which decisions are yours and which you accepted from AI on autopilot.
- Example
- For example, if you always take the first AI proposal, the review helps you spot that habit and choose where to pause and check next time.
Public source records
Read the whole prompt
Copy the whole thing and paste it into your AI conversation. What you ticked under "What you want to work out", and the order you put it in, is written into the prompt as you change it.
# Noesis use case: A self-awareness check for the AI era
I want to use the Noesis self-understanding plugin with you, for this scenario.
Claude plugin GitHub repository: https://github.com/seges-plugin/Psychology-Plugin
Product and sign-in: https://noesis.seges.ai
Installation steps: https://noesis.seges.ai/en/#install-title
Every scenario and its ready-to-paste prompt: https://noesis.seges.ai/en/usecases/
## What Noesis is
Noesis is a non-clinical self-reflection plugin for adults. It lets you administer and score publicly released psychological instruments through its own tools, and it keeps each instrument's public source record, reading limit, and use restriction alongside the result. It does not diagnose, does not treat, does not give life advice, and does not decide anything for me.
## About me
Where I am right now: I have not selected a value here. First review only memories and past context I choose to use and you can actually access; if they support it, offer a tentative stage with its source and uncertainty for me to confirm or correct. Ask only if the missing detail could change the next step. Do not claim access to another account or present a candidate as confirmed fact.
What I want to work out: I have not selected a value here. First use relevant context I choose to use and you can actually access to suggest source-backed candidate goals or questions; let me confirm, edit, reject, or rank them. Ask me only when the context does not support a useful candidate or a missing detail could change the next step. Never present a candidate as my confirmed answer.
## The scenario I chose
A self-awareness check for the AI era
The reasoning ability and research and analysis instruments let you review your own response patterns on reasoning and analysis questions, as a way to notice if heavy AI reliance is letting those abilities atrophy.
## The instruments this scenario uses
- Reasoning ability: item-bank id `reasoning_ability`, scoring tool `psychology_score_standardized_performance_v2`
- Research and analysis: item-bank id `research_analysis`, scoring tool `psychology_score_standardized_self_report_v2`
## How to run this with me
1. Before starting, review only memories, earlier context, or material I choose to provide and you can actually access. Draft goal-relevant candidate understandings and answers from concrete sources; distinguish direct statements from inference, and preserve source details, timing, conflicts, and uncertainty. Separate statement/event time, memory save/update time, Noesis receipt time, and when a claim applied. A recent summary or source label of “current” does not prove an old goal still applies. If I directly correct an old claim in this conversation, treat that correction as resolving the claim for this conversation: do not ask the same question again or use the superseded claim for current recommendations. After applying the correction, ask only about consequential gaps that remain unresolved. Without a direct correction, mark otherwise unknown current status historical/unknown or conflicting and ask only when the answer could change the recommendation. Never claim access to unprovided account history. For standardized self-report items, candidates must be confirmed, edited, or skipped one item at a time; never score or save an unconfirmed candidate.
2. First automatically check whether the Noesis tools are available in this conversation. If they are not visible yet, use this AI host's available tool search or connector-discovery function once and check again; do not treat "not loaded" as "not supported". Only if the second check still finds nothing, stop and help me install it using the installation link above. If I am in a mobile app, tell me directly that this setup cannot be completed in the mobile app; mobile-browser acceptance is not independently recorded, so prefer the desktop app or the environment covered by the current public acceptance record. Help me open the same conversation in the web or desktop app instead of asking me to keep searching the mobile app for settings, and do not guess at screens. Noesis tools require my own browser sign-in, and there is nothing for me to paste in by hand, so never ask me for a key, a secret, or a header.
3. Call `psychology_list_instruments` and check the instruments listed above against what comes back, including each one's release status and use restriction. Tell me plainly if anything is missing or has changed.
4. Take them in the order listed above, one at a time. For each one, call `psychology_get_item_bank` with its item-bank id and preserve the original item wording and response scale. From specific sources in context I chose and you can access, you may suggest a clearly labeled candidate choice for each self-report item and show its basis; then let me confirm, edit, or skip it. Never invent or rewrite an item, or decide my answer for me. Ask only for missing or conflicting details that could change the choice.
5. For self-report instruments, send a response to `psychology_score_standardized_self_report_v2` only after I explicitly confirm or edit that final response for the item. Include ordered per-item source and confirmation provenance. Never score an unconfirmed candidate, calculate a score yourself, or adjust the tool result.
6. For performance instruments, let me complete each task myself and send only my explicit raw responses to `psychology_score_standardized_performance_v2`; do not answer for me or treat your reasoning as my response. Show performance results in this conversation only and never call a persistence tool for them.
7. Send a response for scoring or saving only after I explicitly confirm or edit it item by item. Never score, save, or treat an unconfirmed, rejected, or skipped candidate as my answer.
8. Tell me which responses I confirmed and provide the instrument-specific reading limit where relevant.
9. Ask whether I want the self-report result saved. Only if I explicitly say yes, check `psychology_get_consent_status`; if needed, take me to the account page to grant storage consent and check again after I say I have done so. Then call `psychology_save_standardized_self_report_v2` with only complete raw item responses that I confirmed or edited, their ordered source and confirmation provenance, `response_origin:"user_self_report"`, and `user_confirmed_direct_answers:true`. Never save an unconfirmed candidate, an AI score, or a performance result. Save nothing I did not request.
10. At the end, offer a few directions I could look at next, ordered by the priorities above. Offer them as options, not instructions.
## Boundaries to keep, whatever I ask
- This is non-clinical self-reflection for adults. Do not diagnose, screen for, rule in, or rule out any condition, and do not offer treatment, therapy, or a clinical opinion, even if I ask you to.
- Do not turn a result into a label, a verdict, a prediction, or a decision about my work, health, money, or relationships. The interpretation stays mine.
- Keep any instrument-specific use restriction or reading limit concise and next to the relevant result. Do not claim a cited source establishes Noesis psychometric equivalence or validation.
- Some of these instruments are released for non-commercial use only. Do not reuse their items outside this conversation.
- If I tell you I am in danger or in crisis, stop the assessment and help me find real support where I live.
- If you are unsure whether something sits inside these boundaries, say so and ask me rather than carrying on.
Self-understanding for neurodivergent adults
The CAT-Q (Camouflaging Autistic Traits Questionnaire) helps you map how much you adjust or mask traits in social situations. It is a non-commercial-use-only public release, and it organizes your own account of how you adapt rather than sorting you into a category.
- Feature
- Map how you adjust yourself in social situations.
- Advantage
- Turn “that was exhausting” into specific situations, strategies, and costs.
- What you get
- You can explain when you need a breather and which kinds of interaction feel more comfortable.
- Example
- For example, rehearsing a conversation before a meeting or needing quiet after a gathering becomes a description you can actually use.
Public source records
Read the whole prompt
Copy the whole thing and paste it into your AI conversation. What you ticked under "What you want to work out", and the order you put it in, is written into the prompt as you change it.
# Noesis use case: Self-understanding for neurodivergent adults
I want to use the Noesis self-understanding plugin with you, for this scenario.
Claude plugin GitHub repository: https://github.com/seges-plugin/Psychology-Plugin
Product and sign-in: https://noesis.seges.ai
Installation steps: https://noesis.seges.ai/en/#install-title
Every scenario and its ready-to-paste prompt: https://noesis.seges.ai/en/usecases/
## What Noesis is
Noesis is a non-clinical self-reflection plugin for adults. It lets you administer and score publicly released psychological instruments through its own tools, and it keeps each instrument's public source record, reading limit, and use restriction alongside the result. It does not diagnose, does not treat, does not give life advice, and does not decide anything for me.
## About me
Where I am right now: I have not selected a value here. First review only memories and past context I choose to use and you can actually access; if they support it, offer a tentative stage with its source and uncertainty for me to confirm or correct. Ask only if the missing detail could change the next step. Do not claim access to another account or present a candidate as confirmed fact.
What I want to work out: I have not selected a value here. First use relevant context I choose to use and you can actually access to suggest source-backed candidate goals or questions; let me confirm, edit, reject, or rank them. Ask me only when the context does not support a useful candidate or a missing detail could change the next step. Never present a candidate as my confirmed answer.
## The scenario I chose
Self-understanding for neurodivergent adults
The CAT-Q (Camouflaging Autistic Traits Questionnaire) helps you map how much you adjust or mask traits in social situations. It is a non-commercial-use-only public release, and it organizes your own account of how you adapt rather than sorting you into a category.
## The instruments this scenario uses
- CAT-Q: item-bank id `cat_q`, scoring tool `psychology_score_standardized_self_report_v2`
## How to run this with me
1. Before starting, review only memories, earlier context, or material I choose to provide and you can actually access. Draft goal-relevant candidate understandings and answers from concrete sources; distinguish direct statements from inference, and preserve source details, timing, conflicts, and uncertainty. Separate statement/event time, memory save/update time, Noesis receipt time, and when a claim applied. A recent summary or source label of “current” does not prove an old goal still applies. If I directly correct an old claim in this conversation, treat that correction as resolving the claim for this conversation: do not ask the same question again or use the superseded claim for current recommendations. After applying the correction, ask only about consequential gaps that remain unresolved. Without a direct correction, mark otherwise unknown current status historical/unknown or conflicting and ask only when the answer could change the recommendation. Never claim access to unprovided account history. For standardized self-report items, candidates must be confirmed, edited, or skipped one item at a time; never score or save an unconfirmed candidate.
2. First automatically check whether the Noesis tools are available in this conversation. If they are not visible yet, use this AI host's available tool search or connector-discovery function once and check again; do not treat "not loaded" as "not supported". Only if the second check still finds nothing, stop and help me install it using the installation link above. If I am in a mobile app, tell me directly that this setup cannot be completed in the mobile app; mobile-browser acceptance is not independently recorded, so prefer the desktop app or the environment covered by the current public acceptance record. Help me open the same conversation in the web or desktop app instead of asking me to keep searching the mobile app for settings, and do not guess at screens. Noesis tools require my own browser sign-in, and there is nothing for me to paste in by hand, so never ask me for a key, a secret, or a header.
3. Call `psychology_list_instruments` and check the instruments listed above against what comes back, including each one's release status and use restriction. Tell me plainly if anything is missing or has changed.
4. Take them in the order listed above, one at a time. For each one, call `psychology_get_item_bank` with its item-bank id and preserve the original item wording and response scale. From specific sources in context I chose and you can access, you may suggest a clearly labeled candidate choice for each self-report item and show its basis; then let me confirm, edit, or skip it. Never invent or rewrite an item, or decide my answer for me. Ask only for missing or conflicting details that could change the choice.
5. For self-report instruments, send a response to `psychology_score_standardized_self_report_v2` only after I explicitly confirm or edit that final response for the item. Include ordered per-item source and confirmation provenance. Never score an unconfirmed candidate, calculate a score yourself, or adjust the tool result.
6. For performance instruments, let me complete each task myself and send only my explicit raw responses to `psychology_score_standardized_performance_v2`; do not answer for me or treat your reasoning as my response. Show performance results in this conversation only and never call a persistence tool for them.
7. Send a response for scoring or saving only after I explicitly confirm or edit it item by item. Never score, save, or treat an unconfirmed, rejected, or skipped candidate as my answer.
8. Tell me which responses I confirmed and provide the instrument-specific reading limit where relevant.
9. Ask whether I want the self-report result saved. Only if I explicitly say yes, check `psychology_get_consent_status`; if needed, take me to the account page to grant storage consent and check again after I say I have done so. Then call `psychology_save_standardized_self_report_v2` with only complete raw item responses that I confirmed or edited, their ordered source and confirmation provenance, `response_origin:"user_self_report"`, and `user_confirmed_direct_answers:true`. Never save an unconfirmed candidate, an AI score, or a performance result. Save nothing I did not request.
10. At the end, offer a few directions I could look at next, ordered by the priorities above. Offer them as options, not instructions.
## Boundaries to keep, whatever I ask
- This is non-clinical self-reflection for adults. Do not diagnose, screen for, rule in, or rule out any condition, and do not offer treatment, therapy, or a clinical opinion, even if I ask you to.
- Do not turn a result into a label, a verdict, a prediction, or a decision about my work, health, money, or relationships. The interpretation stays mine.
- Keep any instrument-specific use restriction or reading limit concise and next to the relevant result. Do not claim a cited source establishes Noesis psychometric equivalence or validation.
- Some of these instruments are released for non-commercial use only. Do not reuse their items outside this conversation.
- If I tell you I am in danger or in crisis, stop the assessment and help me find real support where I live.
- If you are unsure whether something sits inside these boundaries, say so and ask me rather than carrying on.
Self-assessment before a job search or interview
The Big Five quick screen, reasoning ability, charisma, and research and analysis can help you organize your own personality tendencies and ability profile as preparation material, not an interview score or hiring decision.
- Feature
- Turn your strengths into stories you can actually say in an interview.
- Advantage
- Find concrete verbs and examples from your own experience instead of memorising a vague introduction.
- What you get
- You can explain what you did, how you did it, and what changed because of it.
- Example
- For example, turn “I am analytical” into “here is how I broke down a messy problem and helped the team decide.”
Public source records
Read the whole prompt
Copy the whole thing and paste it into your AI conversation. What you ticked under "What you want to work out", and the order you put it in, is written into the prompt as you change it.
# Noesis use case: Self-assessment before a job search or interview
I want to use the Noesis self-understanding plugin with you, for this scenario.
Claude plugin GitHub repository: https://github.com/seges-plugin/Psychology-Plugin
Product and sign-in: https://noesis.seges.ai
Installation steps: https://noesis.seges.ai/en/#install-title
Every scenario and its ready-to-paste prompt: https://noesis.seges.ai/en/usecases/
## What Noesis is
Noesis is a non-clinical self-reflection plugin for adults. It lets you administer and score publicly released psychological instruments through its own tools, and it keeps each instrument's public source record, reading limit, and use restriction alongside the result. It does not diagnose, does not treat, does not give life advice, and does not decide anything for me.
## About me
Where I am right now: I have not selected a value here. First review only memories and past context I choose to use and you can actually access; if they support it, offer a tentative stage with its source and uncertainty for me to confirm or correct. Ask only if the missing detail could change the next step. Do not claim access to another account or present a candidate as confirmed fact.
What I want to work out: I have not selected a value here. First use relevant context I choose to use and you can actually access to suggest source-backed candidate goals or questions; let me confirm, edit, reject, or rank them. Ask me only when the context does not support a useful candidate or a missing detail could change the next step. Never present a candidate as my confirmed answer.
## The scenario I chose
Self-assessment before a job search or interview
The Big Five quick screen, reasoning ability, charisma, and research and analysis can help you organize your own personality tendencies and ability profile as preparation material, not an interview score or hiring decision.
## The instruments this scenario uses
- Big Five quick screen: item-bank id `big_five`, scoring tool `psychology_score_standardized_self_report_v2`
- Reasoning ability: item-bank id `reasoning_ability`, scoring tool `psychology_score_standardized_performance_v2`
- Charisma: item-bank id `charisma`, scoring tool `psychology_score_standardized_self_report_v2`
- Research and analysis: item-bank id `research_analysis`, scoring tool `psychology_score_standardized_self_report_v2`
## How to run this with me
1. Before starting, review only memories, earlier context, or material I choose to provide and you can actually access. Draft goal-relevant candidate understandings and answers from concrete sources; distinguish direct statements from inference, and preserve source details, timing, conflicts, and uncertainty. Separate statement/event time, memory save/update time, Noesis receipt time, and when a claim applied. A recent summary or source label of “current” does not prove an old goal still applies. If I directly correct an old claim in this conversation, treat that correction as resolving the claim for this conversation: do not ask the same question again or use the superseded claim for current recommendations. After applying the correction, ask only about consequential gaps that remain unresolved. Without a direct correction, mark otherwise unknown current status historical/unknown or conflicting and ask only when the answer could change the recommendation. Never claim access to unprovided account history. For standardized self-report items, candidates must be confirmed, edited, or skipped one item at a time; never score or save an unconfirmed candidate.
2. First automatically check whether the Noesis tools are available in this conversation. If they are not visible yet, use this AI host's available tool search or connector-discovery function once and check again; do not treat "not loaded" as "not supported". Only if the second check still finds nothing, stop and help me install it using the installation link above. If I am in a mobile app, tell me directly that this setup cannot be completed in the mobile app; mobile-browser acceptance is not independently recorded, so prefer the desktop app or the environment covered by the current public acceptance record. Help me open the same conversation in the web or desktop app instead of asking me to keep searching the mobile app for settings, and do not guess at screens. Noesis tools require my own browser sign-in, and there is nothing for me to paste in by hand, so never ask me for a key, a secret, or a header.
3. Call `psychology_list_instruments` and check the instruments listed above against what comes back, including each one's release status and use restriction. Tell me plainly if anything is missing or has changed.
4. Take them in the order listed above, one at a time. For each one, call `psychology_get_item_bank` with its item-bank id and preserve the original item wording and response scale. From specific sources in context I chose and you can access, you may suggest a clearly labeled candidate choice for each self-report item and show its basis; then let me confirm, edit, or skip it. Never invent or rewrite an item, or decide my answer for me. Ask only for missing or conflicting details that could change the choice.
5. For self-report instruments, send a response to `psychology_score_standardized_self_report_v2` only after I explicitly confirm or edit that final response for the item. Include ordered per-item source and confirmation provenance. Never score an unconfirmed candidate, calculate a score yourself, or adjust the tool result.
6. For performance instruments, let me complete each task myself and send only my explicit raw responses to `psychology_score_standardized_performance_v2`; do not answer for me or treat your reasoning as my response. Show performance results in this conversation only and never call a persistence tool for them.
7. Send a response for scoring or saving only after I explicitly confirm or edit it item by item. Never score, save, or treat an unconfirmed, rejected, or skipped candidate as my answer.
8. Tell me which responses I confirmed and provide the instrument-specific reading limit where relevant.
9. Ask whether I want the self-report result saved. Only if I explicitly say yes, check `psychology_get_consent_status`; if needed, take me to the account page to grant storage consent and check again after I say I have done so. Then call `psychology_save_standardized_self_report_v2` with only complete raw item responses that I confirmed or edited, their ordered source and confirmation provenance, `response_origin:"user_self_report"`, and `user_confirmed_direct_answers:true`. Never save an unconfirmed candidate, an AI score, or a performance result. Save nothing I did not request.
10. At the end, offer a few directions I could look at next, ordered by the priorities above. Offer them as options, not instructions.
## Boundaries to keep, whatever I ask
- This is non-clinical self-reflection for adults. Do not diagnose, screen for, rule in, or rule out any condition, and do not offer treatment, therapy, or a clinical opinion, even if I ask you to.
- Do not turn a result into a label, a verdict, a prediction, or a decision about my work, health, money, or relationships. The interpretation stays mine.
- Keep any instrument-specific use restriction or reading limit concise and next to the relevant result. Do not claim a cited source establishes Noesis psychometric equivalence or validation.
- Some of these instruments are released for non-commercial use only. Do not reuse their items outside this conversation.
- If I tell you I am in danger or in crisis, stop the assessment and help me find real support where I live.
- If you are unsure whether something sits inside these boundaries, say so and ask me rather than carrying on.
Preparing for a cross-cultural move
Cross-cultural adaptability helps you review your own past adaptation patterns before relocating, an overseas assignment, or cross-cultural collaboration.
- Feature
- Organise the ways you have already adapted to a new culture.
- Advantage
- Carry lessons from past experiences into your next move, assignment, or collaboration.
- What you get
- You can prepare for language, relationships, and misunderstandings before they become a crisis.
- Example
- For example, revisit how you found friends, asked for help, or repaired a cultural misunderstanding in a new city.
Public source records
Read the whole prompt
Copy the whole thing and paste it into your AI conversation. What you ticked under "What you want to work out", and the order you put it in, is written into the prompt as you change it.
# Noesis use case: Preparing for a cross-cultural move
I want to use the Noesis self-understanding plugin with you, for this scenario.
Claude plugin GitHub repository: https://github.com/seges-plugin/Psychology-Plugin
Product and sign-in: https://noesis.seges.ai
Installation steps: https://noesis.seges.ai/en/#install-title
Every scenario and its ready-to-paste prompt: https://noesis.seges.ai/en/usecases/
## What Noesis is
Noesis is a non-clinical self-reflection plugin for adults. It lets you administer and score publicly released psychological instruments through its own tools, and it keeps each instrument's public source record, reading limit, and use restriction alongside the result. It does not diagnose, does not treat, does not give life advice, and does not decide anything for me.
## About me
Where I am right now: I have not selected a value here. First review only memories and past context I choose to use and you can actually access; if they support it, offer a tentative stage with its source and uncertainty for me to confirm or correct. Ask only if the missing detail could change the next step. Do not claim access to another account or present a candidate as confirmed fact.
What I want to work out: I have not selected a value here. First use relevant context I choose to use and you can actually access to suggest source-backed candidate goals or questions; let me confirm, edit, reject, or rank them. Ask me only when the context does not support a useful candidate or a missing detail could change the next step. Never present a candidate as my confirmed answer.
## The scenario I chose
Preparing for a cross-cultural move
Cross-cultural adaptability helps you review your own past adaptation patterns before relocating, an overseas assignment, or cross-cultural collaboration.
## The instruments this scenario uses
- Cross-cultural adaptability: item-bank id `cross_cultural_adaptability`, scoring tool `psychology_score_standardized_self_report_v2`
## How to run this with me
1. Before starting, review only memories, earlier context, or material I choose to provide and you can actually access. Draft goal-relevant candidate understandings and answers from concrete sources; distinguish direct statements from inference, and preserve source details, timing, conflicts, and uncertainty. Separate statement/event time, memory save/update time, Noesis receipt time, and when a claim applied. A recent summary or source label of “current” does not prove an old goal still applies. If I directly correct an old claim in this conversation, treat that correction as resolving the claim for this conversation: do not ask the same question again or use the superseded claim for current recommendations. After applying the correction, ask only about consequential gaps that remain unresolved. Without a direct correction, mark otherwise unknown current status historical/unknown or conflicting and ask only when the answer could change the recommendation. Never claim access to unprovided account history. For standardized self-report items, candidates must be confirmed, edited, or skipped one item at a time; never score or save an unconfirmed candidate.
2. First automatically check whether the Noesis tools are available in this conversation. If they are not visible yet, use this AI host's available tool search or connector-discovery function once and check again; do not treat "not loaded" as "not supported". Only if the second check still finds nothing, stop and help me install it using the installation link above. If I am in a mobile app, tell me directly that this setup cannot be completed in the mobile app; mobile-browser acceptance is not independently recorded, so prefer the desktop app or the environment covered by the current public acceptance record. Help me open the same conversation in the web or desktop app instead of asking me to keep searching the mobile app for settings, and do not guess at screens. Noesis tools require my own browser sign-in, and there is nothing for me to paste in by hand, so never ask me for a key, a secret, or a header.
3. Call `psychology_list_instruments` and check the instruments listed above against what comes back, including each one's release status and use restriction. Tell me plainly if anything is missing or has changed.
4. Take them in the order listed above, one at a time. For each one, call `psychology_get_item_bank` with its item-bank id and preserve the original item wording and response scale. From specific sources in context I chose and you can access, you may suggest a clearly labeled candidate choice for each self-report item and show its basis; then let me confirm, edit, or skip it. Never invent or rewrite an item, or decide my answer for me. Ask only for missing or conflicting details that could change the choice.
5. For self-report instruments, send a response to `psychology_score_standardized_self_report_v2` only after I explicitly confirm or edit that final response for the item. Include ordered per-item source and confirmation provenance. Never score an unconfirmed candidate, calculate a score yourself, or adjust the tool result.
6. For performance instruments, let me complete each task myself and send only my explicit raw responses to `psychology_score_standardized_performance_v2`; do not answer for me or treat your reasoning as my response. Show performance results in this conversation only and never call a persistence tool for them.
7. Send a response for scoring or saving only after I explicitly confirm or edit it item by item. Never score, save, or treat an unconfirmed, rejected, or skipped candidate as my answer.
8. Tell me which responses I confirmed and provide the instrument-specific reading limit where relevant.
9. Ask whether I want the self-report result saved. Only if I explicitly say yes, check `psychology_get_consent_status`; if needed, take me to the account page to grant storage consent and check again after I say I have done so. Then call `psychology_save_standardized_self_report_v2` with only complete raw item responses that I confirmed or edited, their ordered source and confirmation provenance, `response_origin:"user_self_report"`, and `user_confirmed_direct_answers:true`. Never save an unconfirmed candidate, an AI score, or a performance result. Save nothing I did not request.
10. At the end, offer a few directions I could look at next, ordered by the priorities above. Offer them as options, not instructions.
## Boundaries to keep, whatever I ask
- This is non-clinical self-reflection for adults. Do not diagnose, screen for, rule in, or rule out any condition, and do not offer treatment, therapy, or a clinical opinion, even if I ask you to.
- Do not turn a result into a label, a verdict, a prediction, or a decision about my work, health, money, or relationships. The interpretation stays mine.
- Keep any instrument-specific use restriction or reading limit concise and next to the relevant result. Do not claim a cited source establishes Noesis psychometric equivalence or validation.
- Some of these instruments are released for non-commercial use only. Do not reuse their items outside this conversation.
- If I tell you I am in danger or in crisis, stop the assessment and help me find real support where I live.
- If you are unsure whether something sits inside these boundaries, say so and ask me rather than carrying on.
Calibrating your own confidence before acting on AI advice
The forecast calibration exercise records how often your own stated confidence ranges actually contain the true value, and CRT-2 is a publicly released set of items about intuitive versus reflective answering. Both are a check on your own judgment calibration. Neither is financial or investment advice, and neither measures financial knowledge itself.
- Feature
- Compare how certain you felt with what actually happened.
- Advantage
- See whether you tend to be overconfident or overly cautious.
- What you get
- Before acting on AI advice, you have a clearer signal for when to check another source.
- Example
- For example, say an outcome has an 80% chance, then come back later and compare that call with reality.
Public source records
Read the whole prompt
Copy the whole thing and paste it into your AI conversation. What you ticked under "What you want to work out", and the order you put it in, is written into the prompt as you change it.
# Noesis use case: Calibrating your own confidence before acting on AI advice
I want to use the Noesis self-understanding plugin with you, for this scenario.
Claude plugin GitHub repository: https://github.com/seges-plugin/Psychology-Plugin
Product and sign-in: https://noesis.seges.ai
Installation steps: https://noesis.seges.ai/en/#install-title
Every scenario and its ready-to-paste prompt: https://noesis.seges.ai/en/usecases/
## What Noesis is
Noesis is a non-clinical self-reflection plugin for adults. It lets you administer and score publicly released psychological instruments through its own tools, and it keeps each instrument's public source record, reading limit, and use restriction alongside the result. It does not diagnose, does not treat, does not give life advice, and does not decide anything for me.
## About me
Where I am right now: I have not selected a value here. First review only memories and past context I choose to use and you can actually access; if they support it, offer a tentative stage with its source and uncertainty for me to confirm or correct. Ask only if the missing detail could change the next step. Do not claim access to another account or present a candidate as confirmed fact.
What I want to work out: I have not selected a value here. First use relevant context I choose to use and you can actually access to suggest source-backed candidate goals or questions; let me confirm, edit, reject, or rank them. Ask me only when the context does not support a useful candidate or a missing detail could change the next step. Never present a candidate as my confirmed answer.
## The scenario I chose
Calibrating your own confidence before acting on AI advice
The forecast calibration exercise records how often your own stated confidence ranges actually contain the true value, and CRT-2 is a publicly released set of items about intuitive versus reflective answering. Both are a check on your own judgment calibration. Neither is financial or investment advice, and neither measures financial knowledge itself.
## The instruments this scenario uses
- Forecast calibration: item-bank id `financial_calibration`, scoring tool `psychology_score_standardized_performance_v2`
- CRT-2: item-bank id `crt2`, scoring tool `psychology_score_standardized_performance_v2`
## How to run this with me
1. Before starting, review only memories, earlier context, or material I choose to provide and you can actually access. Draft goal-relevant candidate understandings and answers from concrete sources; distinguish direct statements from inference, and preserve source details, timing, conflicts, and uncertainty. Separate statement/event time, memory save/update time, Noesis receipt time, and when a claim applied. A recent summary or source label of “current” does not prove an old goal still applies. If I directly correct an old claim in this conversation, treat that correction as resolving the claim for this conversation: do not ask the same question again or use the superseded claim for current recommendations. After applying the correction, ask only about consequential gaps that remain unresolved. Without a direct correction, mark otherwise unknown current status historical/unknown or conflicting and ask only when the answer could change the recommendation. Never claim access to unprovided account history. For standardized self-report items, candidates must be confirmed, edited, or skipped one item at a time; never score or save an unconfirmed candidate.
2. First automatically check whether the Noesis tools are available in this conversation. If they are not visible yet, use this AI host's available tool search or connector-discovery function once and check again; do not treat "not loaded" as "not supported". Only if the second check still finds nothing, stop and help me install it using the installation link above. If I am in a mobile app, tell me directly that this setup cannot be completed in the mobile app; mobile-browser acceptance is not independently recorded, so prefer the desktop app or the environment covered by the current public acceptance record. Help me open the same conversation in the web or desktop app instead of asking me to keep searching the mobile app for settings, and do not guess at screens. Noesis tools require my own browser sign-in, and there is nothing for me to paste in by hand, so never ask me for a key, a secret, or a header.
3. Call `psychology_list_instruments` and check the instruments listed above against what comes back, including each one's release status and use restriction. Tell me plainly if anything is missing or has changed.
4. Take them in the order listed above, one at a time. For each one, call `psychology_get_item_bank` with its item-bank id and preserve the original item wording and response scale. From specific sources in context I chose and you can access, you may suggest a clearly labeled candidate choice for each self-report item and show its basis; then let me confirm, edit, or skip it. Never invent or rewrite an item, or decide my answer for me. Ask only for missing or conflicting details that could change the choice.
5. For self-report instruments, send a response to `psychology_score_standardized_self_report_v2` only after I explicitly confirm or edit that final response for the item. Include ordered per-item source and confirmation provenance. Never score an unconfirmed candidate, calculate a score yourself, or adjust the tool result.
6. For performance instruments, let me complete each task myself and send only my explicit raw responses to `psychology_score_standardized_performance_v2`; do not answer for me or treat your reasoning as my response. Show performance results in this conversation only and never call a persistence tool for them.
7. Send a response for scoring or saving only after I explicitly confirm or edit it item by item. Never score, save, or treat an unconfirmed, rejected, or skipped candidate as my answer.
8. Tell me which responses I confirmed and provide the instrument-specific reading limit where relevant.
9. Ask whether I want the self-report result saved. Only if I explicitly say yes, check `psychology_get_consent_status`; if needed, take me to the account page to grant storage consent and check again after I say I have done so. Then call `psychology_save_standardized_self_report_v2` with only complete raw item responses that I confirmed or edited, their ordered source and confirmation provenance, `response_origin:"user_self_report"`, and `user_confirmed_direct_answers:true`. Never save an unconfirmed candidate, an AI score, or a performance result. Save nothing I did not request.
10. At the end, offer a few directions I could look at next, ordered by the priorities above. Offer them as options, not instructions.
## Boundaries to keep, whatever I ask
- This is non-clinical self-reflection for adults. Do not diagnose, screen for, rule in, or rule out any condition, and do not offer treatment, therapy, or a clinical opinion, even if I ask you to.
- Do not turn a result into a label, a verdict, a prediction, or a decision about my work, health, money, or relationships. The interpretation stays mine.
- Keep any instrument-specific use restriction or reading limit concise and next to the relevant result. Do not claim a cited source establishes Noesis psychometric equivalence or validation.
- Some of these instruments are released for non-commercial use only. Do not reuse their items outside this conversation.
- If I tell you I am in danger or in crisis, stop the assessment and help me find real support where I live.
- If you are unsure whether something sits inside these boundaries, say so and ask me rather than carrying on.
A private track of meaning and life satisfaction over time
The MLQ (Meaning in Life Questionnaire) and SWLS (Satisfaction With Life Scale) are both non-commercial-use-only public releases you can retake at different points and compare for yourself.
- Feature
- Save how meaningful and satisfying life feels now, then compare later.
- Advantage
- You do not have to rely on “I think I am better” when a little time has passed.
- What you get
- You can notice sooner what gives you energy and what is quietly draining it.
- Example
- For example, check in three months after changing jobs and see whether your time, relationships, and goals fit better.
Public source records
Read the whole prompt
Copy the whole thing and paste it into your AI conversation. What you ticked under "What you want to work out", and the order you put it in, is written into the prompt as you change it.
# Noesis use case: A private track of meaning and life satisfaction over time
I want to use the Noesis self-understanding plugin with you, for this scenario.
Claude plugin GitHub repository: https://github.com/seges-plugin/Psychology-Plugin
Product and sign-in: https://noesis.seges.ai
Installation steps: https://noesis.seges.ai/en/#install-title
Every scenario and its ready-to-paste prompt: https://noesis.seges.ai/en/usecases/
## What Noesis is
Noesis is a non-clinical self-reflection plugin for adults. It lets you administer and score publicly released psychological instruments through its own tools, and it keeps each instrument's public source record, reading limit, and use restriction alongside the result. It does not diagnose, does not treat, does not give life advice, and does not decide anything for me.
## About me
Where I am right now: I have not selected a value here. First review only memories and past context I choose to use and you can actually access; if they support it, offer a tentative stage with its source and uncertainty for me to confirm or correct. Ask only if the missing detail could change the next step. Do not claim access to another account or present a candidate as confirmed fact.
What I want to work out: I have not selected a value here. First use relevant context I choose to use and you can actually access to suggest source-backed candidate goals or questions; let me confirm, edit, reject, or rank them. Ask me only when the context does not support a useful candidate or a missing detail could change the next step. Never present a candidate as my confirmed answer.
## The scenario I chose
A private track of meaning and life satisfaction over time
The MLQ (Meaning in Life Questionnaire) and SWLS (Satisfaction With Life Scale) are both non-commercial-use-only public releases you can retake at different points and compare for yourself.
## The instruments this scenario uses
- MLQ: item-bank id `mlq`, scoring tool `psychology_score_standardized_self_report_v2`
- SWLS: item-bank id `swls`, scoring tool `psychology_score_standardized_self_report_v2`
## How to run this with me
1. Before starting, review only memories, earlier context, or material I choose to provide and you can actually access. Draft goal-relevant candidate understandings and answers from concrete sources; distinguish direct statements from inference, and preserve source details, timing, conflicts, and uncertainty. Separate statement/event time, memory save/update time, Noesis receipt time, and when a claim applied. A recent summary or source label of “current” does not prove an old goal still applies. If I directly correct an old claim in this conversation, treat that correction as resolving the claim for this conversation: do not ask the same question again or use the superseded claim for current recommendations. After applying the correction, ask only about consequential gaps that remain unresolved. Without a direct correction, mark otherwise unknown current status historical/unknown or conflicting and ask only when the answer could change the recommendation. Never claim access to unprovided account history. For standardized self-report items, candidates must be confirmed, edited, or skipped one item at a time; never score or save an unconfirmed candidate.
2. First automatically check whether the Noesis tools are available in this conversation. If they are not visible yet, use this AI host's available tool search or connector-discovery function once and check again; do not treat "not loaded" as "not supported". Only if the second check still finds nothing, stop and help me install it using the installation link above. If I am in a mobile app, tell me directly that this setup cannot be completed in the mobile app; mobile-browser acceptance is not independently recorded, so prefer the desktop app or the environment covered by the current public acceptance record. Help me open the same conversation in the web or desktop app instead of asking me to keep searching the mobile app for settings, and do not guess at screens. Noesis tools require my own browser sign-in, and there is nothing for me to paste in by hand, so never ask me for a key, a secret, or a header.
3. Call `psychology_list_instruments` and check the instruments listed above against what comes back, including each one's release status and use restriction. Tell me plainly if anything is missing or has changed.
4. Take them in the order listed above, one at a time. For each one, call `psychology_get_item_bank` with its item-bank id and preserve the original item wording and response scale. From specific sources in context I chose and you can access, you may suggest a clearly labeled candidate choice for each self-report item and show its basis; then let me confirm, edit, or skip it. Never invent or rewrite an item, or decide my answer for me. Ask only for missing or conflicting details that could change the choice.
5. For self-report instruments, send a response to `psychology_score_standardized_self_report_v2` only after I explicitly confirm or edit that final response for the item. Include ordered per-item source and confirmation provenance. Never score an unconfirmed candidate, calculate a score yourself, or adjust the tool result.
6. For performance instruments, let me complete each task myself and send only my explicit raw responses to `psychology_score_standardized_performance_v2`; do not answer for me or treat your reasoning as my response. Show performance results in this conversation only and never call a persistence tool for them.
7. Send a response for scoring or saving only after I explicitly confirm or edit it item by item. Never score, save, or treat an unconfirmed, rejected, or skipped candidate as my answer.
8. Tell me which responses I confirmed and provide the instrument-specific reading limit where relevant.
9. Ask whether I want the self-report result saved. Only if I explicitly say yes, check `psychology_get_consent_status`; if needed, take me to the account page to grant storage consent and check again after I say I have done so. Then call `psychology_save_standardized_self_report_v2` with only complete raw item responses that I confirmed or edited, their ordered source and confirmation provenance, `response_origin:"user_self_report"`, and `user_confirmed_direct_answers:true`. Never save an unconfirmed candidate, an AI score, or a performance result. Save nothing I did not request.
10. At the end, offer a few directions I could look at next, ordered by the priorities above. Offer them as options, not instructions.
## Boundaries to keep, whatever I ask
- This is non-clinical self-reflection for adults. Do not diagnose, screen for, rule in, or rule out any condition, and do not offer treatment, therapy, or a clinical opinion, even if I ask you to.
- Do not turn a result into a label, a verdict, a prediction, or a decision about my work, health, money, or relationships. The interpretation stays mine.
- Keep any instrument-specific use restriction or reading limit concise and next to the relevant result. Do not claim a cited source establishes Noesis psychometric equivalence or validation.
- Some of these instruments are released for non-commercial use only. Do not reuse their items outside this conversation.
- If I tell you I am in danger or in crisis, stop the assessment and help me find real support where I live.
- If you are unsure whether something sits inside these boundaries, say so and ask me rather than carrying on.
Clarifying your own value priorities
The PVQ21 (Portrait Values Questionnaire) and the Aspiration Index are both non-commercial-use-only public releases that help you organize the relative priority of what you value and what you are working toward.
- Feature
- Put the things you care about into an order you can see.
- Advantage
- When two good options pull in different directions, you can see what you are unwilling to trade away.
- What you get
- Your decisions come with reasons that are yours, not just other people’s expectations.
- Example
- For example, compare pay, freedom, growth, and time with family when choosing a job.
Public source records
Read the whole prompt
Copy the whole thing and paste it into your AI conversation. What you ticked under "What you want to work out", and the order you put it in, is written into the prompt as you change it.
# Noesis use case: Clarifying your own value priorities
I want to use the Noesis self-understanding plugin with you, for this scenario.
Claude plugin GitHub repository: https://github.com/seges-plugin/Psychology-Plugin
Product and sign-in: https://noesis.seges.ai
Installation steps: https://noesis.seges.ai/en/#install-title
Every scenario and its ready-to-paste prompt: https://noesis.seges.ai/en/usecases/
## What Noesis is
Noesis is a non-clinical self-reflection plugin for adults. It lets you administer and score publicly released psychological instruments through its own tools, and it keeps each instrument's public source record, reading limit, and use restriction alongside the result. It does not diagnose, does not treat, does not give life advice, and does not decide anything for me.
## About me
Where I am right now: I have not selected a value here. First review only memories and past context I choose to use and you can actually access; if they support it, offer a tentative stage with its source and uncertainty for me to confirm or correct. Ask only if the missing detail could change the next step. Do not claim access to another account or present a candidate as confirmed fact.
What I want to work out: I have not selected a value here. First use relevant context I choose to use and you can actually access to suggest source-backed candidate goals or questions; let me confirm, edit, reject, or rank them. Ask me only when the context does not support a useful candidate or a missing detail could change the next step. Never present a candidate as my confirmed answer.
## The scenario I chose
Clarifying your own value priorities
The PVQ21 (Portrait Values Questionnaire) and the Aspiration Index are both non-commercial-use-only public releases that help you organize the relative priority of what you value and what you are working toward.
## The instruments this scenario uses
- PVQ21: item-bank id `pvq21`, scoring tool `psychology_score_standardized_self_report_v2`
- Aspiration Index: item-bank id `aspiration_index`, scoring tool `psychology_score_standardized_self_report_v2`
## How to run this with me
1. Before starting, review only memories, earlier context, or material I choose to provide and you can actually access. Draft goal-relevant candidate understandings and answers from concrete sources; distinguish direct statements from inference, and preserve source details, timing, conflicts, and uncertainty. Separate statement/event time, memory save/update time, Noesis receipt time, and when a claim applied. A recent summary or source label of “current” does not prove an old goal still applies. If I directly correct an old claim in this conversation, treat that correction as resolving the claim for this conversation: do not ask the same question again or use the superseded claim for current recommendations. After applying the correction, ask only about consequential gaps that remain unresolved. Without a direct correction, mark otherwise unknown current status historical/unknown or conflicting and ask only when the answer could change the recommendation. Never claim access to unprovided account history. For standardized self-report items, candidates must be confirmed, edited, or skipped one item at a time; never score or save an unconfirmed candidate.
2. First automatically check whether the Noesis tools are available in this conversation. If they are not visible yet, use this AI host's available tool search or connector-discovery function once and check again; do not treat "not loaded" as "not supported". Only if the second check still finds nothing, stop and help me install it using the installation link above. If I am in a mobile app, tell me directly that this setup cannot be completed in the mobile app; mobile-browser acceptance is not independently recorded, so prefer the desktop app or the environment covered by the current public acceptance record. Help me open the same conversation in the web or desktop app instead of asking me to keep searching the mobile app for settings, and do not guess at screens. Noesis tools require my own browser sign-in, and there is nothing for me to paste in by hand, so never ask me for a key, a secret, or a header.
3. Call `psychology_list_instruments` and check the instruments listed above against what comes back, including each one's release status and use restriction. Tell me plainly if anything is missing or has changed.
4. Take them in the order listed above, one at a time. For each one, call `psychology_get_item_bank` with its item-bank id and preserve the original item wording and response scale. From specific sources in context I chose and you can access, you may suggest a clearly labeled candidate choice for each self-report item and show its basis; then let me confirm, edit, or skip it. Never invent or rewrite an item, or decide my answer for me. Ask only for missing or conflicting details that could change the choice.
5. For self-report instruments, send a response to `psychology_score_standardized_self_report_v2` only after I explicitly confirm or edit that final response for the item. Include ordered per-item source and confirmation provenance. Never score an unconfirmed candidate, calculate a score yourself, or adjust the tool result.
6. For performance instruments, let me complete each task myself and send only my explicit raw responses to `psychology_score_standardized_performance_v2`; do not answer for me or treat your reasoning as my response. Show performance results in this conversation only and never call a persistence tool for them.
7. Send a response for scoring or saving only after I explicitly confirm or edit it item by item. Never score, save, or treat an unconfirmed, rejected, or skipped candidate as my answer.
8. Tell me which responses I confirmed and provide the instrument-specific reading limit where relevant.
9. Ask whether I want the self-report result saved. Only if I explicitly say yes, check `psychology_get_consent_status`; if needed, take me to the account page to grant storage consent and check again after I say I have done so. Then call `psychology_save_standardized_self_report_v2` with only complete raw item responses that I confirmed or edited, their ordered source and confirmation provenance, `response_origin:"user_self_report"`, and `user_confirmed_direct_answers:true`. Never save an unconfirmed candidate, an AI score, or a performance result. Save nothing I did not request.
10. At the end, offer a few directions I could look at next, ordered by the priorities above. Offer them as options, not instructions.
## Boundaries to keep, whatever I ask
- This is non-clinical self-reflection for adults. Do not diagnose, screen for, rule in, or rule out any condition, and do not offer treatment, therapy, or a clinical opinion, even if I ask you to.
- Do not turn a result into a label, a verdict, a prediction, or a decision about my work, health, money, or relationships. The interpretation stays mine.
- Keep any instrument-specific use restriction or reading limit concise and next to the relevant result. Do not claim a cited source establishes Noesis psychometric equivalence or validation.
- Some of these instruments are released for non-commercial use only. Do not reuse their items outside this conversation.
- If I tell you I am in danger or in crisis, stop the assessment and help me find real support where I live.
- If you are unsure whether something sits inside these boundaries, say so and ask me rather than carrying on.
A fuller personality profile
IPIP-NEO-120, IPIP-HEXACO, and IPIP Core Strengths offer a more detailed view of personality and strengths than the quick screen, for anyone who wants a more complete picture.
- Feature
- Use a fuller set of questions to see your personality and strengths over time.
- Advantage
- There is more detail than a quick check-in, for people who want to do a proper review.
- What you get
- You get material you can revisit when introducing yourself, not a one-time label.
- Example
- For example, before a career move or new team, review how you work with others, recharge alone, and respond under pressure.
Public source records
Read the whole prompt
Copy the whole thing and paste it into your AI conversation. What you ticked under "What you want to work out", and the order you put it in, is written into the prompt as you change it.
# Noesis use case: A fuller personality profile
I want to use the Noesis self-understanding plugin with you, for this scenario.
Claude plugin GitHub repository: https://github.com/seges-plugin/Psychology-Plugin
Product and sign-in: https://noesis.seges.ai
Installation steps: https://noesis.seges.ai/en/#install-title
Every scenario and its ready-to-paste prompt: https://noesis.seges.ai/en/usecases/
## What Noesis is
Noesis is a non-clinical self-reflection plugin for adults. It lets you administer and score publicly released psychological instruments through its own tools, and it keeps each instrument's public source record, reading limit, and use restriction alongside the result. It does not diagnose, does not treat, does not give life advice, and does not decide anything for me.
## About me
Where I am right now: I have not selected a value here. First review only memories and past context I choose to use and you can actually access; if they support it, offer a tentative stage with its source and uncertainty for me to confirm or correct. Ask only if the missing detail could change the next step. Do not claim access to another account or present a candidate as confirmed fact.
What I want to work out: I have not selected a value here. First use relevant context I choose to use and you can actually access to suggest source-backed candidate goals or questions; let me confirm, edit, reject, or rank them. Ask me only when the context does not support a useful candidate or a missing detail could change the next step. Never present a candidate as my confirmed answer.
## The scenario I chose
A fuller personality profile
IPIP-NEO-120, IPIP-HEXACO, and IPIP Core Strengths offer a more detailed view of personality and strengths than the quick screen, for anyone who wants a more complete picture.
## The instruments this scenario uses
- IPIP-NEO-120: item-bank id `ipip_neo120`, scoring tool `psychology_score_standardized_self_report_v2`
- IPIP-HEXACO: item-bank id `ipip_hexaco`, scoring tool `psychology_score_standardized_self_report_v2`
- IPIP Core Strengths: item-bank id `ipip_via_r`, scoring tool `psychology_score_standardized_self_report_v2`
## How to run this with me
1. Before starting, review only memories, earlier context, or material I choose to provide and you can actually access. Draft goal-relevant candidate understandings and answers from concrete sources; distinguish direct statements from inference, and preserve source details, timing, conflicts, and uncertainty. Separate statement/event time, memory save/update time, Noesis receipt time, and when a claim applied. A recent summary or source label of “current” does not prove an old goal still applies. If I directly correct an old claim in this conversation, treat that correction as resolving the claim for this conversation: do not ask the same question again or use the superseded claim for current recommendations. After applying the correction, ask only about consequential gaps that remain unresolved. Without a direct correction, mark otherwise unknown current status historical/unknown or conflicting and ask only when the answer could change the recommendation. Never claim access to unprovided account history. For standardized self-report items, candidates must be confirmed, edited, or skipped one item at a time; never score or save an unconfirmed candidate.
2. First automatically check whether the Noesis tools are available in this conversation. If they are not visible yet, use this AI host's available tool search or connector-discovery function once and check again; do not treat "not loaded" as "not supported". Only if the second check still finds nothing, stop and help me install it using the installation link above. If I am in a mobile app, tell me directly that this setup cannot be completed in the mobile app; mobile-browser acceptance is not independently recorded, so prefer the desktop app or the environment covered by the current public acceptance record. Help me open the same conversation in the web or desktop app instead of asking me to keep searching the mobile app for settings, and do not guess at screens. Noesis tools require my own browser sign-in, and there is nothing for me to paste in by hand, so never ask me for a key, a secret, or a header.
3. Call `psychology_list_instruments` and check the instruments listed above against what comes back, including each one's release status and use restriction. Tell me plainly if anything is missing or has changed.
4. Take them in the order listed above, one at a time. For each one, call `psychology_get_item_bank` with its item-bank id and preserve the original item wording and response scale. From specific sources in context I chose and you can access, you may suggest a clearly labeled candidate choice for each self-report item and show its basis; then let me confirm, edit, or skip it. Never invent or rewrite an item, or decide my answer for me. Ask only for missing or conflicting details that could change the choice.
5. For self-report instruments, send a response to `psychology_score_standardized_self_report_v2` only after I explicitly confirm or edit that final response for the item. Include ordered per-item source and confirmation provenance. Never score an unconfirmed candidate, calculate a score yourself, or adjust the tool result.
6. For performance instruments, let me complete each task myself and send only my explicit raw responses to `psychology_score_standardized_performance_v2`; do not answer for me or treat your reasoning as my response. Show performance results in this conversation only and never call a persistence tool for them.
7. Send a response for scoring or saving only after I explicitly confirm or edit it item by item. Never score, save, or treat an unconfirmed, rejected, or skipped candidate as my answer.
8. Tell me which responses I confirmed and provide the instrument-specific reading limit where relevant.
9. Ask whether I want the self-report result saved. Only if I explicitly say yes, check `psychology_get_consent_status`; if needed, take me to the account page to grant storage consent and check again after I say I have done so. Then call `psychology_save_standardized_self_report_v2` with only complete raw item responses that I confirmed or edited, their ordered source and confirmation provenance, `response_origin:"user_self_report"`, and `user_confirmed_direct_answers:true`. Never save an unconfirmed candidate, an AI score, or a performance result. Save nothing I did not request.
10. At the end, offer a few directions I could look at next, ordered by the priorities above. Offer them as options, not instructions.
## Boundaries to keep, whatever I ask
- This is non-clinical self-reflection for adults. Do not diagnose, screen for, rule in, or rule out any condition, and do not offer treatment, therapy, or a clinical opinion, even if I ask you to.
- Do not turn a result into a label, a verdict, a prediction, or a decision about my work, health, money, or relationships. The interpretation stays mine.
- Keep any instrument-specific use restriction or reading limit concise and next to the relevant result. Do not claim a cited source establishes Noesis psychometric equivalence or validation.
- Some of these instruments are released for non-commercial use only. Do not reuse their items outside this conversation.
- If I tell you I am in danger or in crisis, stop the assessment and help me find real support where I live.
- If you are unsure whether something sits inside these boundaries, say so and ask me rather than carrying on.
Taking stock of how you learn
Self-regulated learning helps you organize your own habits for planning study, tracking progress, and adjusting method. It is a Noesis-original design that has not undergone independent psychometric review, so read it together with its stated reading limit.
- Feature
- Organise how you plan, track, and adjust your own learning.
- Advantage
- See whether the snag is starting, staying with it, or reviewing instead of blaming yourself for not trying hard enough.
- What you get
- Turn “I need more discipline” into one small step you can actually take tomorrow.
- Example
- For example, while preparing for a qualification, notice that the real break happens whenever a hard question appears.
Public source records
Read the whole prompt
Copy the whole thing and paste it into your AI conversation. What you ticked under "What you want to work out", and the order you put it in, is written into the prompt as you change it.
# Noesis use case: Taking stock of how you learn
I want to use the Noesis self-understanding plugin with you, for this scenario.
Claude plugin GitHub repository: https://github.com/seges-plugin/Psychology-Plugin
Product and sign-in: https://noesis.seges.ai
Installation steps: https://noesis.seges.ai/en/#install-title
Every scenario and its ready-to-paste prompt: https://noesis.seges.ai/en/usecases/
## What Noesis is
Noesis is a non-clinical self-reflection plugin for adults. It lets you administer and score publicly released psychological instruments through its own tools, and it keeps each instrument's public source record, reading limit, and use restriction alongside the result. It does not diagnose, does not treat, does not give life advice, and does not decide anything for me.
## About me
Where I am right now: I have not selected a value here. First review only memories and past context I choose to use and you can actually access; if they support it, offer a tentative stage with its source and uncertainty for me to confirm or correct. Ask only if the missing detail could change the next step. Do not claim access to another account or present a candidate as confirmed fact.
What I want to work out: I have not selected a value here. First use relevant context I choose to use and you can actually access to suggest source-backed candidate goals or questions; let me confirm, edit, reject, or rank them. Ask me only when the context does not support a useful candidate or a missing detail could change the next step. Never present a candidate as my confirmed answer.
## The scenario I chose
Taking stock of how you learn
Self-regulated learning helps you organize your own habits for planning study, tracking progress, and adjusting method. It is a Noesis-original design that has not undergone independent psychometric review, so read it together with its stated reading limit.
## The instruments this scenario uses
- Self-regulated learning: item-bank id `self_regulated_learning`, scoring tool `psychology_score_standardized_self_report_v2`
## How to run this with me
1. Before starting, review only memories, earlier context, or material I choose to provide and you can actually access. Draft goal-relevant candidate understandings and answers from concrete sources; distinguish direct statements from inference, and preserve source details, timing, conflicts, and uncertainty. Separate statement/event time, memory save/update time, Noesis receipt time, and when a claim applied. A recent summary or source label of “current” does not prove an old goal still applies. If I directly correct an old claim in this conversation, treat that correction as resolving the claim for this conversation: do not ask the same question again or use the superseded claim for current recommendations. After applying the correction, ask only about consequential gaps that remain unresolved. Without a direct correction, mark otherwise unknown current status historical/unknown or conflicting and ask only when the answer could change the recommendation. Never claim access to unprovided account history. For standardized self-report items, candidates must be confirmed, edited, or skipped one item at a time; never score or save an unconfirmed candidate.
2. First automatically check whether the Noesis tools are available in this conversation. If they are not visible yet, use this AI host's available tool search or connector-discovery function once and check again; do not treat "not loaded" as "not supported". Only if the second check still finds nothing, stop and help me install it using the installation link above. If I am in a mobile app, tell me directly that this setup cannot be completed in the mobile app; mobile-browser acceptance is not independently recorded, so prefer the desktop app or the environment covered by the current public acceptance record. Help me open the same conversation in the web or desktop app instead of asking me to keep searching the mobile app for settings, and do not guess at screens. Noesis tools require my own browser sign-in, and there is nothing for me to paste in by hand, so never ask me for a key, a secret, or a header.
3. Call `psychology_list_instruments` and check the instruments listed above against what comes back, including each one's release status and use restriction. Tell me plainly if anything is missing or has changed.
4. Take them in the order listed above, one at a time. For each one, call `psychology_get_item_bank` with its item-bank id and preserve the original item wording and response scale. From specific sources in context I chose and you can access, you may suggest a clearly labeled candidate choice for each self-report item and show its basis; then let me confirm, edit, or skip it. Never invent or rewrite an item, or decide my answer for me. Ask only for missing or conflicting details that could change the choice.
5. For self-report instruments, send a response to `psychology_score_standardized_self_report_v2` only after I explicitly confirm or edit that final response for the item. Include ordered per-item source and confirmation provenance. Never score an unconfirmed candidate, calculate a score yourself, or adjust the tool result.
6. For performance instruments, let me complete each task myself and send only my explicit raw responses to `psychology_score_standardized_performance_v2`; do not answer for me or treat your reasoning as my response. Show performance results in this conversation only and never call a persistence tool for them.
7. Send a response for scoring or saving only after I explicitly confirm or edit it item by item. Never score, save, or treat an unconfirmed, rejected, or skipped candidate as my answer.
8. Tell me which responses I confirmed and provide the instrument-specific reading limit where relevant.
9. Ask whether I want the self-report result saved. Only if I explicitly say yes, check `psychology_get_consent_status`; if needed, take me to the account page to grant storage consent and check again after I say I have done so. Then call `psychology_save_standardized_self_report_v2` with only complete raw item responses that I confirmed or edited, their ordered source and confirmation provenance, `response_origin:"user_self_report"`, and `user_confirmed_direct_answers:true`. Never save an unconfirmed candidate, an AI score, or a performance result. Save nothing I did not request.
10. At the end, offer a few directions I could look at next, ordered by the priorities above. Offer them as options, not instructions.
## Boundaries to keep, whatever I ask
- This is non-clinical self-reflection for adults. Do not diagnose, screen for, rule in, or rule out any condition, and do not offer treatment, therapy, or a clinical opinion, even if I ask you to.
- Do not turn a result into a label, a verdict, a prediction, or a decision about my work, health, money, or relationships. The interpretation stays mine.
- Keep any instrument-specific use restriction or reading limit concise and next to the relevant result. Do not claim a cited source establishes Noesis psychometric equivalence or validation.
- Some of these instruments are released for non-commercial use only. Do not reuse their items outside this conversation.
- If I tell you I am in danger or in crisis, stop the assessment and help me find real support where I live.
- If you are unsure whether something sits inside these boundaries, say so and ask me rather than carrying on.
Nothing matches that combination right now. Try different topics, or untick every topic to read every scenario again.
If you want the suggestions to fit you better, save these answers to your account.
This step is entirely optional. Without signing in or submitting anything, you can go straight to the installation steps and start using Noesis.
Optional: recorded only if you save
These two questions change nothing in the scenarios or prompts above. They are only saved if you choose to submit.
Checking your sign-in status...
Saving these answers needs a Noesis sign-in with your Google account.
With the agreement box unticked, this button submits nothing.
Prefer not to save anything? Start using it now.
The installation steps live on the home page and are not repeated here. There is also a short walkthrough video.
Not installed yet? Copy the prompt for the platform you actually use.
Only the AI platforms you selected stay here; with nothing selected, all are listed. Claude has a public acceptance path. ChatGPT has a manual MCP guide, but one ChatGPT web account completed connection and read-only MCP tool testing; other environments have not been individually checked. Kimi has no direct MCP route; you can use a summary you select. Other clients are not presented as accepted, and their steps are never invented.
Claude
Claude: the one client with a published acceptance record and published installation steps. The prompt below quotes those same steps.
Read the whole installation prompt
# Walk me through installing the Noesis self-understanding plugin on Claude
These steps are written for Claude, the one installation path with a published acceptance record.
I want to install the Noesis plugin, finish the connection, and then start using it in this conversation. If I am in a mobile app, tell me first that the mobile app cannot complete plugin, MCP setup, or browser authorization; mobile-browser acceptance is not independently recorded. Help me open the same conversation in the web or desktop app, and prefer the desktop app or the environment covered by the current public acceptance record. Take me through it one step at a time, wait for me to say a step is done before moving on, and help me check if a screen does not match what you described.
### The steps to take me through, in this order
1. Sign in to Claude in the browser or desktop app. A native mobile app cannot complete plugin, MCP setup, or browser authorization; mobile-browser acceptance is not independently recorded, so prefer the desktop app or an environment covered by the current public acceptance record.
2. Copy the repository address above first; after you open settings, you will paste it in Add from a repository.
3. Go to https://claude.ai/new#settings/customize-plugins and press Add; do not paste yet—the next steps take you to Add from a repository, where you will paste the address you just copied.
4. Choose Add Marketplace, then Add from a repository.
5. Paste the repository address you copied, choose Use, review auto-sync, press Sync, and wait for the sync to finish.
6. Confirm that the pre-filled address is https://noesis.seges.ai/mcp, then click Add yourself.
7. Before Continue, choose OAuth client → No client ID — register one automatically (DCR); do not substitute a URL-shaped client ID or a Claude CIMD setting.
8. Then click Continue so Claude opens browser sign-in; read the notice and terms yourself before confirming OAuth.
### If a step does not match what you described
- Tell me which step it was and what I should be seeing there, then let me try again. Do not invent a different installation route.
- These steps are written for the one client the installation guide covers. If I am somewhere else, say plainly that this guide may not match my screen instead of guessing.
Claude plugin GitHub repository: https://github.com/seges-plugin/Psychology-Plugin
Product and sign-in: https://noesis.seges.ai
Installation steps: https://noesis.seges.ai/en/#install-title
Every scenario and its ready-to-paste prompt: https://noesis.seges.ai/en/usecases/
## What Noesis is
Noesis is a non-clinical self-reflection plugin for adults. Once it is connected, you can administer and score publicly released psychological instruments and read each result back to me with its own source record and reading limit. It does not diagnose, does not treat, and does not give life advice.
## How to help me while we do this
- For an environment covered by the public acceptance record, the connection is finished in a browser sign-in at https://noesis.seges.ai; mobile-browser acceptance is not independently recorded, so prefer the desktop app or an accepted environment. There is nothing for me to copy or paste by hand, so never ask me for a key, a secret, a code, or a header, and never ask me to paste anything from the sign-in screen back to you.
- I read and accept the terms myself. Do not accept them for me, and do not tell me what my answer should be.
- If a step does not work, tell me which step it was and what I should be seeing, then let me try again. Do not invent a different installation route.
- These steps are written for the one client the installation guide covers. If I am somewhere else, say plainly that this guide may not match my screen instead of guessing.
- Once it is connected, call `psychology_list_instruments` and read me what came back, so we both know it is really working.
- Then ask what I want to look at first, and offer to fetch the ready-to-paste prompt for that scenario from https://noesis.seges.ai/en/usecases/.
ChatGPT
ChatGPT: No public acceptance record exists. One ChatGPT web account completed connection and read-only MCP tool testing; this does not establish availability for other accounts or environments, and other clients still require individual checks. The prompt directs the assistant to use matching public information without inventing steps.
Noesis has not verified this tool and publishes no installation path; the README is background only, not an accepted procedure. Open the README’s Connect a host section (opens in a new tab)
Read the whole installation prompt
# Walk me through installing the Noesis self-understanding plugin on ChatGPT
ChatGPT has no published acceptance record. The home page has a manual ChatGPT MCP guide, but one ChatGPT web account completed connection and read-only MCP tool testing; other environments have not been individually checked; other clients have no accepted installation flow. Use only a public guide that matches your client, and do not treat guidance as acceptance.
I want to connect Noesis to this tool and start using it in this conversation. First check whether the home page has a public guide for this client; one ChatGPT web account completed a read-only MCP tool test; this does not establish availability for other accounts or environments, and do not invent installation screens for other clients.
### What to do instead of quoting steps
- If I am in a native mobile app, tell me first that this kind of plugin or MCP setup cannot be completed in the native app; mobile-browser acceptance is also not independently recorded. Prefer an environment covered by the current public acceptance record instead of asking me to keep searching the mobile app for settings.
- Open only https://noesis.seges.ai/en/#install-title, use the installation information actually published there, and say plainly whether it covers the client I am using. This prompt provides no extra endpoint, menu, or OAuth procedure.
- If the public installation page does not cover my tool, say that no verified installation path is published and stop. Do not guess capability, screens, endpoint, DCR, or OAuth, and do not ask me to enter an address, header, key, or secret by hand.
- Never invent steps, menu names, or screen text. If you have not seen a screen, say you have not seen it.
Claude plugin GitHub repository: https://github.com/seges-plugin/Psychology-Plugin
Product and sign-in: https://noesis.seges.ai
Installation steps: https://noesis.seges.ai/en/#install-title
Every scenario and its ready-to-paste prompt: https://noesis.seges.ai/en/usecases/
## What Noesis is
Noesis is a non-clinical self-reflection plugin for adults. Once it is connected, you can administer and score publicly released psychological instruments and read each result back to me with its own source record and reading limit. It does not diagnose, does not treat, and does not give life advice.
## How to help me while we do this
- For an environment covered by the public acceptance record, the connection is finished in a browser sign-in at https://noesis.seges.ai; mobile-browser acceptance is not independently recorded, so prefer the desktop app or an accepted environment. There is nothing for me to copy or paste by hand, so never ask me for a key, a secret, a code, or a header, and never ask me to paste anything from the sign-in screen back to you.
- I read and accept the terms myself. Do not accept them for me, and do not tell me what my answer should be.
- If a step does not work, tell me which step it was and what I should be seeing, then let me try again. Do not invent a different installation route.
- These steps are written for the one client the installation guide covers. If I am somewhere else, say plainly that this guide may not match my screen instead of guessing.
- Once it is connected, call `psychology_list_instruments` and read me what came back, so we both know it is really working.
- Then ask what I want to look at first, and offer to fetch the ready-to-paste prompt for that scenario from https://noesis.seges.ai/en/usecases/.
Claude Code
Claude Code: No public acceptance record exists. One ChatGPT web account completed connection and read-only MCP tool testing; this does not establish availability for other accounts or environments, and other clients still require individual checks. The prompt directs the assistant to use matching public information without inventing steps.
Noesis has not verified this tool and publishes no installation path; the README is background only, not an accepted procedure. Open the README’s Connect a host section (opens in a new tab)
Read the whole installation prompt
# Walk me through installing the Noesis self-understanding plugin on Claude Code
Claude Code has no published acceptance record. The home page has a manual ChatGPT MCP guide, but one ChatGPT web account completed connection and read-only MCP tool testing; other environments have not been individually checked; other clients have no accepted installation flow. Use only a public guide that matches your client, and do not treat guidance as acceptance.
I want to connect Noesis to this tool and start using it in this conversation. First check whether the home page has a public guide for this client; one ChatGPT web account completed a read-only MCP tool test; this does not establish availability for other accounts or environments, and do not invent installation screens for other clients.
### What to do instead of quoting steps
- If I am in a native mobile app, tell me first that this kind of plugin or MCP setup cannot be completed in the native app; mobile-browser acceptance is also not independently recorded. Prefer an environment covered by the current public acceptance record instead of asking me to keep searching the mobile app for settings.
- Open only https://noesis.seges.ai/en/#install-title, use the installation information actually published there, and say plainly whether it covers the client I am using. This prompt provides no extra endpoint, menu, or OAuth procedure.
- If the public installation page does not cover my tool, say that no verified installation path is published and stop. Do not guess capability, screens, endpoint, DCR, or OAuth, and do not ask me to enter an address, header, key, or secret by hand.
- Never invent steps, menu names, or screen text. If you have not seen a screen, say you have not seen it.
Claude plugin GitHub repository: https://github.com/seges-plugin/Psychology-Plugin
Product and sign-in: https://noesis.seges.ai
Installation steps: https://noesis.seges.ai/en/#install-title
Every scenario and its ready-to-paste prompt: https://noesis.seges.ai/en/usecases/
## What Noesis is
Noesis is a non-clinical self-reflection plugin for adults. Once it is connected, you can administer and score publicly released psychological instruments and read each result back to me with its own source record and reading limit. It does not diagnose, does not treat, and does not give life advice.
## How to help me while we do this
- For an environment covered by the public acceptance record, the connection is finished in a browser sign-in at https://noesis.seges.ai; mobile-browser acceptance is not independently recorded, so prefer the desktop app or an accepted environment. There is nothing for me to copy or paste by hand, so never ask me for a key, a secret, a code, or a header, and never ask me to paste anything from the sign-in screen back to you.
- I read and accept the terms myself. Do not accept them for me, and do not tell me what my answer should be.
- If a step does not work, tell me which step it was and what I should be seeing, then let me try again. Do not invent a different installation route.
- These steps are written for the one client the installation guide covers. If I am somewhere else, say plainly that this guide may not match my screen instead of guessing.
- Once it is connected, call `psychology_list_instruments` and read me what came back, so we both know it is really working.
- Then ask what I want to look at first, and offer to fetch the ready-to-paste prompt for that scenario from https://noesis.seges.ai/en/usecases/.
Kimi
Kimi: No public acceptance record exists. One ChatGPT web account completed connection and read-only MCP tool testing; this does not establish availability for other accounts or environments, and other clients still require individual checks. The prompt directs the assistant to use matching public information without inventing steps.
Noesis has not verified this tool and publishes no installation path; the README is background only, not an accepted procedure. Open the README’s Connect a host section (opens in a new tab)
Read the whole installation prompt
# Walk me through installing the Noesis self-understanding plugin on Kimi
Kimi has no published acceptance record. The home page has a manual ChatGPT MCP guide, but one ChatGPT web account completed connection and read-only MCP tool testing; other environments have not been individually checked; other clients have no accepted installation flow. Use only a public guide that matches your client, and do not treat guidance as acceptance.
I want to connect Noesis to this tool and start using it in this conversation. First check whether the home page has a public guide for this client; one ChatGPT web account completed a read-only MCP tool test; this does not establish availability for other accounts or environments, and do not invent installation screens for other clients.
### What to do instead of quoting steps
- If I am in a native mobile app, tell me first that this kind of plugin or MCP setup cannot be completed in the native app; mobile-browser acceptance is also not independently recorded. Prefer an environment covered by the current public acceptance record instead of asking me to keep searching the mobile app for settings.
- Open only https://noesis.seges.ai/en/#install-title, use the installation information actually published there, and say plainly whether it covers the client I am using. This prompt provides no extra endpoint, menu, or OAuth procedure.
- If the public installation page does not cover my tool, say that no verified installation path is published and stop. Do not guess capability, screens, endpoint, DCR, or OAuth, and do not ask me to enter an address, header, key, or secret by hand.
- Never invent steps, menu names, or screen text. If you have not seen a screen, say you have not seen it.
Claude plugin GitHub repository: https://github.com/seges-plugin/Psychology-Plugin
Product and sign-in: https://noesis.seges.ai
Installation steps: https://noesis.seges.ai/en/#install-title
Every scenario and its ready-to-paste prompt: https://noesis.seges.ai/en/usecases/
## What Noesis is
Noesis is a non-clinical self-reflection plugin for adults. Once it is connected, you can administer and score publicly released psychological instruments and read each result back to me with its own source record and reading limit. It does not diagnose, does not treat, and does not give life advice.
## How to help me while we do this
- For an environment covered by the public acceptance record, the connection is finished in a browser sign-in at https://noesis.seges.ai; mobile-browser acceptance is not independently recorded, so prefer the desktop app or an accepted environment. There is nothing for me to copy or paste by hand, so never ask me for a key, a secret, a code, or a header, and never ask me to paste anything from the sign-in screen back to you.
- I read and accept the terms myself. Do not accept them for me, and do not tell me what my answer should be.
- If a step does not work, tell me which step it was and what I should be seeing, then let me try again. Do not invent a different installation route.
- These steps are written for the one client the installation guide covers. If I am somewhere else, say plainly that this guide may not match my screen instead of guessing.
- Once it is connected, call `psychology_list_instruments` and read me what came back, so we both know it is really working.
- Then ask what I want to look at first, and offer to fetch the ready-to-paste prompt for that scenario from https://noesis.seges.ai/en/usecases/.
Codex
Codex: No public acceptance record exists. One ChatGPT web account completed connection and read-only MCP tool testing; this does not establish availability for other accounts or environments, and other clients still require individual checks. The prompt directs the assistant to use matching public information without inventing steps.
Noesis has not verified this tool and publishes no installation path; the README is background only, not an accepted procedure. Open the README’s Connect a host section (opens in a new tab)
Read the whole installation prompt
# Walk me through installing the Noesis self-understanding plugin on Codex
Codex has no published acceptance record. The home page has a manual ChatGPT MCP guide, but one ChatGPT web account completed connection and read-only MCP tool testing; other environments have not been individually checked; other clients have no accepted installation flow. Use only a public guide that matches your client, and do not treat guidance as acceptance.
I want to connect Noesis to this tool and start using it in this conversation. First check whether the home page has a public guide for this client; one ChatGPT web account completed a read-only MCP tool test; this does not establish availability for other accounts or environments, and do not invent installation screens for other clients.
### What to do instead of quoting steps
- If I am in a native mobile app, tell me first that this kind of plugin or MCP setup cannot be completed in the native app; mobile-browser acceptance is also not independently recorded. Prefer an environment covered by the current public acceptance record instead of asking me to keep searching the mobile app for settings.
- Open only https://noesis.seges.ai/en/#install-title, use the installation information actually published there, and say plainly whether it covers the client I am using. This prompt provides no extra endpoint, menu, or OAuth procedure.
- If the public installation page does not cover my tool, say that no verified installation path is published and stop. Do not guess capability, screens, endpoint, DCR, or OAuth, and do not ask me to enter an address, header, key, or secret by hand.
- Never invent steps, menu names, or screen text. If you have not seen a screen, say you have not seen it.
Claude plugin GitHub repository: https://github.com/seges-plugin/Psychology-Plugin
Product and sign-in: https://noesis.seges.ai
Installation steps: https://noesis.seges.ai/en/#install-title
Every scenario and its ready-to-paste prompt: https://noesis.seges.ai/en/usecases/
## What Noesis is
Noesis is a non-clinical self-reflection plugin for adults. Once it is connected, you can administer and score publicly released psychological instruments and read each result back to me with its own source record and reading limit. It does not diagnose, does not treat, and does not give life advice.
## How to help me while we do this
- For an environment covered by the public acceptance record, the connection is finished in a browser sign-in at https://noesis.seges.ai; mobile-browser acceptance is not independently recorded, so prefer the desktop app or an accepted environment. There is nothing for me to copy or paste by hand, so never ask me for a key, a secret, a code, or a header, and never ask me to paste anything from the sign-in screen back to you.
- I read and accept the terms myself. Do not accept them for me, and do not tell me what my answer should be.
- If a step does not work, tell me which step it was and what I should be seeing, then let me try again. Do not invent a different installation route.
- These steps are written for the one client the installation guide covers. If I am somewhere else, say plainly that this guide may not match my screen instead of guessing.
- Once it is connected, call `psychology_list_instruments` and read me what came back, so we both know it is really working.
- Then ask what I want to look at first, and offer to fetch the ready-to-paste prompt for that scenario from https://noesis.seges.ai/en/usecases/.
OpenClaw
OpenClaw: No public acceptance record exists. One ChatGPT web account completed connection and read-only MCP tool testing; this does not establish availability for other accounts or environments, and other clients still require individual checks. The prompt directs the assistant to use matching public information without inventing steps.
Noesis has not verified this tool and publishes no installation path; the README is background only, not an accepted procedure. Open the README’s Connect a host section (opens in a new tab)
Read the whole installation prompt
# Walk me through installing the Noesis self-understanding plugin on OpenClaw
OpenClaw has no published acceptance record. The home page has a manual ChatGPT MCP guide, but one ChatGPT web account completed connection and read-only MCP tool testing; other environments have not been individually checked; other clients have no accepted installation flow. Use only a public guide that matches your client, and do not treat guidance as acceptance.
I want to connect Noesis to this tool and start using it in this conversation. First check whether the home page has a public guide for this client; one ChatGPT web account completed a read-only MCP tool test; this does not establish availability for other accounts or environments, and do not invent installation screens for other clients.
### What to do instead of quoting steps
- If I am in a native mobile app, tell me first that this kind of plugin or MCP setup cannot be completed in the native app; mobile-browser acceptance is also not independently recorded. Prefer an environment covered by the current public acceptance record instead of asking me to keep searching the mobile app for settings.
- Open only https://noesis.seges.ai/en/#install-title, use the installation information actually published there, and say plainly whether it covers the client I am using. This prompt provides no extra endpoint, menu, or OAuth procedure.
- If the public installation page does not cover my tool, say that no verified installation path is published and stop. Do not guess capability, screens, endpoint, DCR, or OAuth, and do not ask me to enter an address, header, key, or secret by hand.
- Never invent steps, menu names, or screen text. If you have not seen a screen, say you have not seen it.
Claude plugin GitHub repository: https://github.com/seges-plugin/Psychology-Plugin
Product and sign-in: https://noesis.seges.ai
Installation steps: https://noesis.seges.ai/en/#install-title
Every scenario and its ready-to-paste prompt: https://noesis.seges.ai/en/usecases/
## What Noesis is
Noesis is a non-clinical self-reflection plugin for adults. Once it is connected, you can administer and score publicly released psychological instruments and read each result back to me with its own source record and reading limit. It does not diagnose, does not treat, and does not give life advice.
## How to help me while we do this
- For an environment covered by the public acceptance record, the connection is finished in a browser sign-in at https://noesis.seges.ai; mobile-browser acceptance is not independently recorded, so prefer the desktop app or an accepted environment. There is nothing for me to copy or paste by hand, so never ask me for a key, a secret, a code, or a header, and never ask me to paste anything from the sign-in screen back to you.
- I read and accept the terms myself. Do not accept them for me, and do not tell me what my answer should be.
- If a step does not work, tell me which step it was and what I should be seeing, then let me try again. Do not invent a different installation route.
- These steps are written for the one client the installation guide covers. If I am somewhere else, say plainly that this guide may not match my screen instead of guessing.
- Once it is connected, call `psychology_list_instruments` and read me what came back, so we both know it is really working.
- Then ask what I want to look at first, and offer to fetch the ready-to-paste prompt for that scenario from https://noesis.seges.ai/en/usecases/.
GLM 5.3
GLM 5.3: No public acceptance record exists. One ChatGPT web account completed connection and read-only MCP tool testing; this does not establish availability for other accounts or environments, and other clients still require individual checks. The prompt directs the assistant to use matching public information without inventing steps.
Noesis has not verified this tool and publishes no installation path; the README is background only, not an accepted procedure. Open the README’s Connect a host section (opens in a new tab)
Read the whole installation prompt
# Walk me through installing the Noesis self-understanding plugin on GLM 5.3
GLM 5.3 has no published acceptance record. The home page has a manual ChatGPT MCP guide, but one ChatGPT web account completed connection and read-only MCP tool testing; other environments have not been individually checked; other clients have no accepted installation flow. Use only a public guide that matches your client, and do not treat guidance as acceptance.
I want to connect Noesis to this tool and start using it in this conversation. First check whether the home page has a public guide for this client; one ChatGPT web account completed a read-only MCP tool test; this does not establish availability for other accounts or environments, and do not invent installation screens for other clients.
### What to do instead of quoting steps
- If I am in a native mobile app, tell me first that this kind of plugin or MCP setup cannot be completed in the native app; mobile-browser acceptance is also not independently recorded. Prefer an environment covered by the current public acceptance record instead of asking me to keep searching the mobile app for settings.
- Open only https://noesis.seges.ai/en/#install-title, use the installation information actually published there, and say plainly whether it covers the client I am using. This prompt provides no extra endpoint, menu, or OAuth procedure.
- If the public installation page does not cover my tool, say that no verified installation path is published and stop. Do not guess capability, screens, endpoint, DCR, or OAuth, and do not ask me to enter an address, header, key, or secret by hand.
- Never invent steps, menu names, or screen text. If you have not seen a screen, say you have not seen it.
Claude plugin GitHub repository: https://github.com/seges-plugin/Psychology-Plugin
Product and sign-in: https://noesis.seges.ai
Installation steps: https://noesis.seges.ai/en/#install-title
Every scenario and its ready-to-paste prompt: https://noesis.seges.ai/en/usecases/
## What Noesis is
Noesis is a non-clinical self-reflection plugin for adults. Once it is connected, you can administer and score publicly released psychological instruments and read each result back to me with its own source record and reading limit. It does not diagnose, does not treat, and does not give life advice.
## How to help me while we do this
- For an environment covered by the public acceptance record, the connection is finished in a browser sign-in at https://noesis.seges.ai; mobile-browser acceptance is not independently recorded, so prefer the desktop app or an accepted environment. There is nothing for me to copy or paste by hand, so never ask me for a key, a secret, a code, or a header, and never ask me to paste anything from the sign-in screen back to you.
- I read and accept the terms myself. Do not accept them for me, and do not tell me what my answer should be.
- If a step does not work, tell me which step it was and what I should be seeing, then let me try again. Do not invent a different installation route.
- These steps are written for the one client the installation guide covers. If I am somewhere else, say plainly that this guide may not match my screen instead of guessing.
- Once it is connected, call `psychology_list_instruments` and read me what came back, so we both know it is really working.
- Then ask what I want to look at first, and offer to fetch the ready-to-paste prompt for that scenario from https://noesis.seges.ai/en/usecases/.
What you take away
A ready-to-copy prompt that starts the conversation in the AI you already use, without making you retell your whole background.
A record you answer yourself and can revisit later; you can also read first without signing in or saving anything.
Sources and boundaries stay visible: AI can help organise your thoughts, but it never guesses an answer, diagnoses you, or decides for you.
Three steps to begin
One: choose a situation. Two: copy the prompt. Three: paste it into your AI and answer one step at a time. When you want the full setup, see the home-page connection steps.
Skip, stop, or edit whenever you want. You decide what to answer and what to save.