Uncover your team’s unwritten rules.

What the AI will do

• Gather and examine your confirmed team’s interactions, keeping whole conversations together so responses and decisions retain their context.
• Look for recurring unwritten rules, checking examples, exceptions, alternative explanations, and gaps in the record.
• Explain the team’s strengths and sources of communication or collaboration friction, then revise the account with your context.

What you can provide

• Available messages, calendars, and especially meeting transcripts, which preserve how discussions unfold. The AI will investigate access and sharing options.
• Specific accounts of recent interactions when records are unavailable or incomplete; rough examples are welcome.

Where you'll take part

• Confirm or correct the team roster before the investigation.
• Supply missing context and review which patterns ring true or need a different interpretation.

What you'll get

You’ll have an account of your team’s best-supported unwritten rules, strengths, and sources of friction, with examples, exceptions, and evidence limits; if the record is too thin for recurring rules, the AI will explain what it can establish and what is missing.

Activity context

A team’s official values tell you how it hopes to work. Its everyday exchanges reveal the rules people have learned to follow. Have AI examine those exchanges to uncover what helps the team—and what quietly gets in its way.

For the user

Paste this prompt into your work AI. It will establish which team you mean and look for emails, company chat, calendars, and especially meeting transcripts. Transcripts preserve the back-and-forth: whose ideas get attention, how people handle disagreement, and how decisions take shape. That makes them particularly useful for spotting rules nobody states outright. Share any useful records it cannot reach. If records are unavailable, it will interview you about specific team interactions.

You’ll confirm the team, supply missing context, and review its account of your team’s unwritten rules, strengths, and sources of friction.

Guidance for the AI

These are general instructions, not a description of this user's particular AI system, tools, access, or circumstances. Adapt the method to the user's goal and the capabilities you can actually use, while preserving the activity's essential sequence, review points, and intended result.

Start with the activity and the user's context

Read the complete activity and guide the user through it. Use the activity prompt yourself; the user does not need to paste it again. Keep the responsibilities of the user, the AI, and any colleague clear.

Begin with what the user has already supplied. Establish the intended result and resolve missing context through focused questions when needed. Work within the activity's stated scope, permissions, required confirmations, and stopping points, including any deliberate order for introducing information. Make progress with the available information and reasonable, clearly identified assumptions where the activity allows them.

Check the capabilities needed for this work

Inspect relevant available tools, connectors, and data. Use small, relevant, non-destructive checks to establish whether necessary access or a capability actually works. Naming a product does not establish access to it. Let this activity determine what needs checking; do not impose a universal preliminary questionnaire or a full system audit.

When useful access is missing:

• Explain what you cannot reach and how that material would help.
• Investigate practical setup, connection, export, or other access alternatives before asking the user to gather material or troubleshoot.
• If progress depends on access, request only the specific user action or permission needed to enable it or provide the material another way. Continue useful independent work while awaiting the response.
• Offer to walk the user through an export or upload. Welcome files, links, copied text, rough notes, or an interview as supported by your system and the activity, and use the best available route. If the missing access is optional, continue with what is available.

Ask focused questions and respect the answers

Ask questions only when necessary to complete the work well or satisfy the activity's required review points. Build on answers already supplied. Ask one question at a time, or a small group of closely related questions. Wait on work that depends on the answer, and continue useful work that does not. Explain unfamiliar language with examples where helpful.

Confirm answers or decisions you supply on the user's behalf before treating them as the user's final answers or decisions. This does not require separate approval of every sourced finding; follow the activity's specific review requirements.

Keep evidence and independent contributions distinct

Distinguish evidence, inference, and missing information. Check your work against the source material.

Use independent agents when your system actually supports delegation and independent investigation or review would improve the work:

• Give each agent a bounded brief and the context appropriate to its role.
• Keep discovery independent until synthesis where the activity requires it. Preserve any requirement for the user or colleagues to contribute before seeing other explanations.
• Follow the activity's stated fallback and any required confirmation when independent agents or fresh contexts are unavailable. Preserve reviewer isolation where required; describe shared-context or sequential passes accurately.

Produce, check, and revise the result

Do the substantive investigation and production. Carry the activity through to the most complete, usable result within its scope, rather than ending with a plan or handing the main work back to the user. When missing information does not prevent useful work, proceed with clearly identified assumptions where the activity permits them.

Check the result against the activity's completion criteria. At each substantive delivery, state what is complete and any material unfinished work, assumptions, helpful missing information, and remaining uncertainty. Explain how these affect what the user can use, and help the user revise the result.

Preserve the activity's stopping point. Do not turn a rehearsal, investigation, or test-design activity into execution that the user has not authorized.

Your AI prompt

Help me understand how my team works together and the unwritten rules people seem to follow. Investigate its actual interactions, then give me an evidence-based account of its strengths and patterns that may weaken communication, engagement, or collaboration. Ask one focused question at a time when you need my input.

1. Establish the team and gather its interactions.

Use available context to identify the team and its members. If the team is unclear, ask me which one I mean. Show me the roster you propose to examine and wait for me to confirm or correct it.

Check and try available connections to email, company chat, calendars, and meeting transcripts. Gather a broad working collection across recent weeks, recurring meetings, different projects, and different team members. Prioritize meeting transcripts and search for them explicitly. If useful access or transcripts are missing, investigate available connection, transcript-location, and export options before asking me for material. Explain any specific setup action needed from me, the value of read-only access, and offer to draft an IT request if company setup is needed.

If transcripts remain unavailable, explain briefly why they would be particularly valuable for understanding my team. Calendars show planned meetings, and written messages show only part of the exchange. Transcripts can reveal how a discussion actually develops: who introduces an idea, who gets a response, how disagreement is handled, whether someone’s contribution is taken up or passed over, and how the group arrives at a decision. Those sequences can expose unwritten rules that a meeting summary or list of decisions leaves out. Then ask once whether I can share transcripts, offering a workable sharing route from the options you investigated. If I cannot, continue with other available records and interview me about specific meeting exchanges.

Accept copied threads, transcripts, calendar details, and rough examples without requiring a polished collection. Tell me which people, channels, and period the available record covers; extend it where a narrow sample would distort the picture.

When records are unavailable or leave important gaps, interview me about specific recent interactions. Ask about an idea that met resistance, a request that got a quick response, a discussion that stalled, or a meeting that went unusually well. Find out what was said, how others responded, and what happened afterward. Ask for exceptions as well as typical experiences. Treat these accounts as my perspective, not as direct observation of the whole team.

2. Trace the patterns of communication and teamwork.

If separate agents are available and the record is broad enough to benefit, delegate distinct sets of meetings, projects, or channels for parallel examination. Give each agent the confirmed team roster, the scope and dates of its material, and the investigation questions below. Keep whole conversations together so replies and decisions retain their context. Ask each agent to return concrete interaction sequences, possible recurring patterns, counterexamples, and source references. Have them label gaps and tentative interpretations.

As the lead AI, compare their findings across the collection, check the supporting exchanges, resolve conflicting readings where possible, and distinguish team-wide patterns from habits found only in one setting.

Different reports of the same exchange count as one example.

If agents are unavailable or the collection is small, do this work directly.

Read full exchanges with enough context to understand the request or situation. Follow ideas and requests from their introduction through responses, decisions, and follow-through wherever the record allows. Compare repeated instances across members, seniority levels, channels, and ordinary versus urgent work. Notice whose contributions receive attention and what happens when people disagree.

Explore questions such as:

• Do people respond quickly or slowly, and does that vary with who asks or what they need?

• Do they encourage, discourage, challenge, or push one another?

• When a big idea fails to win immediate support, is it directly challenged, explored, or quietly ignored?

• How often does the team meet, how long do meetings run, and who attends or skips them?

• Do meetings lead to decisions and follow-through, or leave issues unresolved?

These are starting points; look for other patterns the evidence reveals.

Interpret each signal in context. A response delay may reflect working hours, urgency, or a move to another channel. A calendar invitation alone does not establish attendance. Check subsequent discussion before treating an unanswered idea as abandoned. Judge meeting frequency and length by what the team needs to accomplish and what happens afterward. Where you cannot check an explanation that matters, ask me or keep the interpretation tentative.

3. Turn recurring patterns into candidate unwritten rules.

For each strong pattern, state the implicit expectation in plain language: for example, “Challenge is welcome in private, but public meetings are for agreement,” or “An idea needs a senior sponsor before anyone develops it.” These are illustrative possibilities, not conclusions to fit the team into.

Explain what repeated behavior supports the rule and what practical effect it appears to have. Include concrete examples with dates and links or other locating details. Look for counterexamples and situations where the rule changes. Keep directly stated team rules separate from the unwritten expectations you infer. Avoid claiming to know people’s private motives. An isolated incident can be a question worth exploring; it is not enough to establish a team norm.

Before presenting the analysis, check that it reflects the team rather than only its loudest voices or best-documented meetings. Where evidence is thin, narrow the finding to the people or situations actually observed. If the material cannot support recurring rules, say what it does show and ask for the specific examples needed to investigate further.

If separate agents are available, give a fresh reviewer the proposed rules, the supporting exchanges, counterexamples, and the known limits of the collection, without your drafting conversation. Ask it to identify claims that outrun the evidence, alternative explanations, and contradictions you have missed. Have it check the source material directly where accessible and identify anything it cannot verify. Use its findings to recheck the evidence and revise or narrow the rules before presenting them. If an independent reviewer is unavailable, perform this check yourself without describing it as independent review.

4. Present the team’s working culture and refine it with me.

Start with a clear overview of how this team appears to communicate and collaborate. Then present its best-supported unwritten rules, each with examples, exceptions, and the effect it seems to have.

Give a list of strengths and a separate list of areas that may lead to weaker communication, engagement, or collaboration. Connect each to the observed patterns. A norm may help in one situation and hinder in another; explain that when it matters. Distinguish demonstrated consequences from plausible effects that we would need to check.

Ask what I recognize and what context would change your reading. Wait for my response, then revise the analysis, making clear which conclusions changed and why.

Done when

You have an account of your team’s unwritten rules, the examples supporting them, its strengths, and patterns that may hinder communication or collaboration, revised with your context.

These instructions were written in September 2026 using the frontier models available at the time. You may now have better tools or models, or restrictions that make some tasks harder. Your AI should adapt to the tools and access you actually have and find the best available way to help. Everything here is a starting suggestion for you and your AI, not a proven best practice for every setup. Feel free to rewrite any of it to better meet your needs.