agentleFS
Baba01hacker666Baba01hacker666/dummyskills/template/SKILL.md

skill-template

Skill0 starsChanged 2 months agoMITView on GitHub

Reusable authoring template for building new dummyskills.

12 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

14 rules. 0 are common, 11 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 6 edited versions we read keep these.

  • 11

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If the task is X, use alternative-skill-1 instead.

    line 22
  2. Very specific:

    If the task is Y, use alternative-skill-2 instead.

    line 23
  3. Very specific:

    If input_one is missing, stop immediately and ask: "Please provide input_one."

    line 32
  4. Distinctive:

    Do not invent missing data.

    line 33
  5. Distinctive:

    Do not assume default values unless explicitly listed here.

    line 34
  6. Distinctive:

    Follow these numbered, ordered steps:

    line 37
  7. Distinctive:

    Read the provided input.

    line 39
  8. Distinctive:

    Follow specific instructions without skipping steps.

    line 42
  9. Distinctive:

    Use explicit "if X, do Y" branches:

    line 47
  10. Distinctive:

    If condition A is met, execute branch A.

    line 48
  11. Distinctive:

    If condition B is met, execute branch B.

    line 49
  12. Distinctive:

    If an unexpected error occurs, follow Section 10 (Common Failures and Recovery).

    line 50
  13. Distinctive:

    Before returning your answer, verify every item:

    line 62
  14. Distinctive:

    Stop work and escalate to the user when:

    line 75

“Common” is measured: kept by at least 6 in 10 of the 6 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

Skill Name

Brief one-paragraph summary of what this skill accomplishes. Use plain English and short sentences.

1. Purpose

State the exact goal of the skill. Define any unfamiliar terms immediately.

2. When to Use

List clear triggers and situations where this skill should be selected:

  • Trigger condition 1
  • Trigger condition 2

3. When Not to Use

List boundaries and alternative skills to avoid misrouting:

  • If the task is X, use alternative-skill-1 instead.
  • If the task is Y, use alternative-skill-2 instead.

4. Required Inputs

List every input required for execution:

  • input_one: Description of what it is and acceptable format.
  • input_two: Description of what it is and acceptable format.

5. Missing-Input Behavior

State exact rules when inputs are missing:

  1. If input_one is missing, stop immediately and ask: "Please provide input_one."
  2. Do not invent missing data.
  3. Do not assume default values unless explicitly listed here.

6. Procedure

Follow these numbered, ordered steps:

  1. Step 1: Inspect and Prepare
    • Read the provided input.
    • Confirm required fields exist.
  2. Step 2: Execute Core Operation
    • Follow specific instructions without skipping steps.
  3. Step 3: Self-Check Output
    • Review against the verification checklist before returning.

7. Decision Rules

Use explicit "if X, do Y" branches:

  • If condition A is met, execute branch A.
  • If condition B is met, execute branch B.
  • If an unexpected error occurs, follow Section 10 (Common Failures and Recovery).

8. Output Format

Specify the exact schema or template of the expected output:

### Title / Summary
- Status: [SUCCESS | BLOCKED | FAILED]
- Result: [Concise description]
- Evidence: [Factual data points]

9. Verification Checklist

Before returning your answer, verify every item:

  • Required input was validated.
  • No facts or data were invented.
  • Output conforms to the required format.
  • All step actions were executed in sequence.

10. Common Failures and Recovery

Failure SymptomCauseRecovery Action
Symptom 1Root cause 1Concrete action to resolve
Symptom 2Root cause 2Concrete action to resolve

11. Stop or Escalation Conditions

Stop work and escalate to the user when:

  • Required input cannot be determined from context.
  • Retry limit is reached (maximum 2 attempts).
  • An operation requires destructive filesystem or network changes without approval.

12. Examples

Example 1: Successful Execution

Input:

[Input payload here]

Expected Observable Actions:

  1. Action 1 taken.
  2. Action 2 taken.

Output:

[Concrete output conforming to Output Format]

Example 2: Missing Input or Failure Case

Input:

[Incomplete or ambiguous input payload here]

Output:

[Exact question or escalation message returned]

Example 3: Expected Final Output

Output:

[Sample canonical completed response]

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.