agentleFS
vbonkvbonk/repo-template/.claude/skills/cot/SKILL.md

cot

Skill8 starsChanged 3 months agoMITView on GitHub

Chain-of-thought analysis before action. Forces structured reasoning, risk assessment, and explicit plan before execution. Use for any complex, destructive, or important request. Invoke with "/cot [request]" or "use cot: [request]". Triggers on "cot", "chain of thought", "think through".

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    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 you want explicit reasoning visible

    line 15
  2. Distinctive:

    When this skill is invoked, analyze the request using this structure:

    line 20
  3. Distinctive:

    Do NOT proceed to execution.

    line 41

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Chain-of-Thought (COT) Skill

Force structured reasoning before action. Prevents rushing into execution.

When to Use

  • Any destructive operation (delete, remove, refactor)
  • Complex multi-step tasks
  • Decisions with significant consequences
  • When you want explicit reasoning visible
  • Any time "think carefully" would normally be said

The Pattern

When this skill is invoked, analyze the request using this structure:

1. What is being asked

Restate the request in clear terms. What is the actual goal?

2. What are the considerations

  • What files/systems are involved?
  • What are the dependencies?
  • What context matters?
  • Are there multiple valid approaches?

3. What could go wrong

  • What are the risks?
  • What's irreversible?
  • What assumptions are being made?
  • What information is missing?

4. Step-by-step plan

Numbered list of discrete actions to accomplish the request.

5. STOP for approval

Do NOT proceed to execution. Wait for explicit:

  • "proceed"
  • "approved"
  • "go ahead"
  • "yes"

Invocation

/cot remove all deprecated API endpoints

use cot: refactor the authentication module

/cot what's the best way to restructure this database schema

Example Output

**COT:**
- **Asked:** Remove deprecated API endpoints from the codebase
- **Considerations:**
  - Need to identify which endpoints are deprecated
  - Check for internal consumers
  - May need migration path for external users
  - Tests reference these endpoints
- **Risk:**
  - Breaking changes for API consumers
  - Dead code might still be imported elsewhere
- **Plan:**
  1. Grep for @deprecated annotations
  2. List all deprecated endpoints with their consumers
  3. Check for external API documentation references
  4. Create removal plan with each file
  5. Stop for approval before any deletion

Awaiting approval to proceed.

Why This Exists

"Think carefully" and "plan thoroughly" don't change Claude's behavior. Concrete output requirements do.

This skill creates a mandatory structured output before any action, making the reasoning visible and creating an approval gate.


Origin: 2025-12-29 session where files were deleted without analysis despite explicit request to "think hard, plan carefully"

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.