AgentStack
SKILL verified MIT Self-run

Docs

skill-pvnarp-agent-skills-docs · by pvnarp

Generates and reviews technical documentation including API docs, architecture decision records, runbooks, and inline code documentation. Use when writing docs, reviewing doc coverage, or creating onboarding materials.

No reviews yet
0 installs
7 views
0.0% view→install

Install

$ agentstack add skill-pvnarp-agent-skills-docs

✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access Used
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

Are you the author of Docs? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Documentation

Documentation has one purpose: help someone do something they couldn't do without it. If it doesn't serve that purpose, delete it.

What to Document (and What Not To)

Document

  • Why decisions were made (not just what was decided)
  • How to get started, deploy, troubleshoot
  • Contracts - API schemas, data formats, integration points
  • Non-obvious behavior - gotchas, workarounds, constraints
  • Runbooks - step-by-step for recurring operational tasks

Don't Document

  • What the code already says (no // increment counter for counter++)
  • Implementation details that change frequently (they'll be wrong tomorrow)
  • Things that can be auto-generated (API docs from schemas, type docs from types)
  • Obvious patterns established by the codebase

Documentation Types

Architecture Decision Record (ADR)

# ADR-001: [Decision Title]

## Status
Accepted | Deprecated | Superseded by ADR-XXX

## Context
[What situation or problem prompted this decision?]

## Decision
[What was decided?]

## Consequences
[What are the positive and negative outcomes?]

Keep ADRs in docs/adr/ or decisions/. Number sequentially. Never delete - mark deprecated.

Runbook

# [Operation Name]

## When to Use
[Trigger condition - when would someone need this?]

## Prerequisites
- [Access, tools, permissions needed]

## Steps
1. [Step with exact command]
2. [Step with expected output]
3. [Step with verification]

## Rollback
[How to undo if something goes wrong]

## Troubleshooting
| Symptom | Cause | Fix |
|---------|-------|-----|

API Documentation

  • Auto-generate from code/schemas where possible (OpenAPI, GraphQL introspection)
  • Document: endpoints, auth, request/response examples, error codes, rate limits
  • Include runnable examples - curl commands, not just schema definitions
  • Keep in sync with code (CI check if possible)

README

Every project needs exactly one README with:

  1. What this is (one sentence)
  2. How to run it locally
  3. How to run tests
  4. How to deploy
  5. Where to find more docs

That's it. Don't put architecture, API docs, or tutorials in the README.

Documentation Review Checklist

  • [ ] Accurate - matches current behavior
  • [ ] Actionable - reader can follow steps and succeed
  • [ ] Findable - in an expected location with a clear name
  • [ ] Minimal - no unnecessary sections or boilerplate
  • [ ] Maintained - has an owner, reviewed when code changes

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet — be the first.

Versions

  • v0.1.0 Imported from the upstream source.