agentleFS
Sign inStart free
nuttakulsvnuttakulsv/think-like/skills/socrates/SKILL.md

socrates

Skill2 starsChanged 41 days agoMITView on GitHub

Socratic questioning before building: ask short, pointed questions that expose undefined terms, hidden assumptions, and contradictions in a request, until the real goal is clear. Use when a request is ambiguous, when requirements conflict, when the agent is about to guess at what the user means, during requirement gathering, or to help someone reason through their own design without handing them the answer (mentoring, interviews). Triggers: "think like Socrates", "ask me questions first", "clarify requirements", "Socratic", "help me think it through", "คิดแบบโสเครตีส", "ถามก่อนทำ", "ช่วยตั้งคำถาม".

10 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use brainstorm.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • nicolasestremSkill

    brainstorm

    Structured idea generation and solution exploration with systematic analysis.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 8 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Summarize the agreed definitions: before building, and keep them in the spec or code comments.

    line 38
  2. Very specific:

    After answers: Agreed definitions: term → meaning, and Out of scope: what was ruled out

    line 45
  3. Distinctive:

    Write in the user's language.

    line 46

“Common” is measured: kept by at least 6 in 10 of the 8 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Think like Socrates

Athenian philosopher known only through others' writing, mostly Plato's dialogues. His method was to ask someone to define what they claimed to know, then ask further questions until the definition either held up or showed its contradictions. Applied to software, it finds the requirement behind the request before any code is written.

When to use

  • A request has words that could mean several things ("admin", "real-time", "user", "report")
  • Requirements from different people contradict each other
  • The agent is about to fill gaps with guesses
  • Mentoring: helping someone find the flaw in their design themselves

Method

  1. Find the key terms in the request and ask for a definition of each that matters: "When you say user, do you mean a customer, a staff member, or both?"
  2. Ask for examples and counter-examples: "Give me one case this must handle, and one it must not."
  3. Test the definition against a case that pushes on it: "If a staff member buys something, are they a user here?" Contradictions mean the definition needs work.
  4. Surface the purpose: "What will you do with the result?" The answer often changes what should be built.
  5. Ask few questions, in order of impact. At most 3–5 at a time, each one answerable in a sentence, the most decision-changing first. Offer a default for each so the user can just confirm.
  6. Summarize the agreed definitions before building, and keep them in the spec or code comments.

In mentoring mode, don't give the answer; ask the next question that leads the person to it.

Output

  • Questions: numbered, highest impact first, each with a proposed default
  • After answers: Agreed definitions: term → meaning, and Out of scope: what was ruled out
  • Write in the user's language. Keep questions short.

Anti-patterns

  • A list of 15 questions before doing anything
  • Questions that could be answered by reading the code or docs (look first)
  • Guessing silently instead of asking about a decision that changes the design
  • Leading questions that hide an opinion

Example (illustration)

User: Add a real-time sales report for admins.

Questions:

  1. "Real-time": within seconds, or is every 5 minutes fine? (Default: 5 minutes, which avoids a streaming setup.)
  2. "Admins": store managers per branch, or head office only? (Default: per branch, seeing only their branch.)
  3. What decision will they make from it during the day? (Default: restock, so per-product quantities matter more than revenue.)

Agreed definitions (after answers): refresh every 5 min; branch managers see their branch, head office sees all; show quantity sold and stock left per product. Out of scope: revenue charts.

Sources

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free