docs-pesap
Skill1 starsChanged 3 months agoMITView on GitHub
Write the shortest complete path that lets the intended reader achieve a real outcome and verify it. Treat implementation, public APIs, configuration, tests, CI, and established project terminology as the source of truth. Split conflicting modes into linked pages instead of making one page serve all four. 1. Identify the reader, desired outcome, document shape, and success check. 2. Inspect the implementation, public contract, configuration, scripts, tests, CI, existing terminology, and related docs before writing. 3. Verify every command, flag,…
In this repository14 other files it gives its agents.
- audit-pi-sessionSkill
- bash-scriptSkill
- cli-uxSkill
- code-quality-gateSkill
- external-reviewSkill
- githubSkill
- gitlabSkill
- good-apiSkill
- prekSkill
- pythonSkill
- rust-pesapSkill
- tdd-pesapSkill
- torcSkill
- uvSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 3 are distinctive, and 1 is 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.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 13
Write the shortest complete path that lets the intended reader achieve a real outcome and verify it.
- Distinctive: line 34
Verify every command, flag, path, version, output, and behavior claimed.
- Distinctive: line 35
Write prerequisites before actions and place validation immediately after the action it verifies.
- Very specific: line 37
Use typed, copy-pasteable examples with realistic placeholders.
≈ “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.
Files
Documentation
Write the shortest complete path that lets the intended reader achieve a real outcome and verify it. Treat implementation, public APIs, configuration, tests, CI, and established project terminology as the source of truth.
Choose the document shape
| Reader need | Shape | Keep it focused on |
| Learn by completing a guided example | Tutorial | A narrow path to a working result |
| Accomplish a known task | How-to | Prerequisites, actions, verification, recovery |
| Look up exact behavior | Reference | Contracts, options, defaults, errors, compatibility |
| Understand concepts or trade-offs | Explanation | Rationale, architecture, and behavior |
Split conflicting modes into linked pages instead of making one page serve all four.
Core workflow
- Identify the reader, desired outcome, document shape, and success check.
- Inspect the implementation, public contract, configuration, scripts, tests, CI, existing terminology, and related docs before writing.
- Verify every command, flag, path, version, output, and behavior claimed.
- Write prerequisites before actions and place validation immediately after the action it verifies. Include recovery where failure is plausible.
- Use typed, copy-pasteable examples with realistic placeholders. Do not invent
APIs or filenames; prefix expected shell output with
#when it shares a command block. - Validate links, anchors, examples, commands, generated references, rendering, and repository documentation checks where available.
Label unverifiable assumptions or omit them. Never present stale or uncertain instructions as authoritative.
Reference router
Open only the reference whose trigger matches the task:
- README guidance — project entry points, quickstarts, success checks, and the learning ladder.
- API documentation — public contracts, errors, side effects, limits, and generated references.
- GitHub Markdown — headings, links, alerts, collapsed sections, tables, diagrams, badges, and accessible rendering.
- Documentation testing — executable examples, vocabulary guards, fixtures, and docs CI.
Completion gate
Confirm that the reader can complete the task in one pass; claims and examples match source truth; prerequisites, permissions, side effects, and recovery are explicit; links and rendering validate; and no filler, marketing claims, hidden steps, or unsupported certainty remain.
Output
Report files and sections changed, reader/outcome/document shape, source-of-truth files consulted, validation commands and results, and any remaining assumptions or documentation gaps.
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
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.

