agentleFS
isac7722isac7722/skills/restate/SKILL.md

restate

Skill1 starsChanged 5 months agoMITView on GitHub

request comprehension verification

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When user invokes this skill: DO NOT execute the original request.

    line 12
  2. Distinctive:

    CRITICAL: ALWAYS respond to the user in the same language being used in the current conversation when presenting your restatement analysis.

    line 16
  3. Distinctive:

    If request references "this file", "that module" → Check conversation for what was discussed

    line 45
  4. Distinctive:

    If request uses pronouns ("it", "that", "them") → Resolve from recent messages

    line 46
  5. Distinctive:

    If request seems incomplete or assumes prior context → Look for setup in previous turns

    line 47
  6. Distinctive:

    Read or modify files

    line 63

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Restate: Request Comprehension Verification

Activation Principle: Leverage expert requirement engineering knowledge to surface hidden assumptions and clarify ambiguities in user requests.

Role

When user invokes this skill: DO NOT execute the original request. Instead, perform meta-analysis and restate their request to verify mutual understanding.

Output Language

CRITICAL: ALWAYS respond to the user in the same language being used in the current conversation when presenting your restatement analysis. Match the user's language choice throughout the restatement.

Restatement Structure

Present your analysis in this 4-part structure:

1. Core Request

"Your request: [one-sentence summary]"

2. My Understanding

  • Goal: What are you trying to achieve?
  • Scope: Where/what does this apply to?
  • Constraints: Any special conditions to consider?
  • Assumptions: [implicit assumptions I'm inferring]

3. Unclear Aspects

[List anything ambiguous as specific questions]

4. Confirmation

"Is my understanding correct?"

Context Boundaries

Available Context:

  • User's message immediately before the /restate command
  • Conversation history (last 5 turns) for resolving implicit references
  • Any files, code, or topics mentioned in the original request

How to Use Context:

  • If request references "this file", "that module" → Check conversation for what was discussed
  • If request uses pronouns ("it", "that", "them") → Resolve from recent messages
  • If request seems incomplete or assumes prior context → Look for setup in previous turns

Example:

User: "Let's refactor the authentication module"
User: "Add better error handling"
User: /restate

→ Understand "Add better error handling" refers to "the authentication module"

Execution Rules

PROHIBITED Actions:

  • Execute the original request
  • Start any work
  • Read or modify files
  • Run any commands

REQUIRED Actions:

  • Analyze only, do not execute
  • Surface implicit assumptions explicitly
  • Convert ambiguities into specific questions
  • Present output in the user's language

Success/Failure Criteria

SUCCESS Indicators:

  • User confirms with "yes", "correct", "that's right"
  • At least 1 hidden assumption surfaced
  • Ambiguous aspects converted to concrete questions

FAILURE Signals:

  • User responds "that's not what I meant"
  • You introduced concepts not in the original request
  • You started execution without confirmation

Error Handling

  • No prior request found → Respond: "There is no request to restate. Please provide a request first."
  • Request is already crystal clear → Still perform restatement, but note: "This request is already very clear"
  • Request is too vague to analyze → Ask: "This request is too brief. Could you provide more details?"

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.