crew-docs
Keep CHANGELOG.md, README.md, SECURITY.md, TODO.md, ADRs, generated docs, and the API and feature reference current as work lands.
Skill by mbadali25
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/mbadali25-crew-docs
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: documents, backend, security, productivity
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/crew-docs
What's inside
- Skill name
crew-docs- Description
- Keep CHANGELOG.md, README.md, SECURITY.md, TODO.md, ADRs, generated docs, and the API and feature reference current as work lands. Use when the user says update the changelog, keep the docs in sync, update the README, maintain TODO, document the API, list the endpoints or features, or asks which docs a change should touch.
- License
- GPL-2.0
- Instructions
- 181 lines in SKILL.md · view on GitHub
Files (2)
- SKILL.md
- integrations.md
SKILL.md (first lines)
# Document maintenance
## The default is: do not touch
Updating every document on every change is how documentation becomes noise. A
CHANGELOG with an entry for each typo fix is unreadable; a README rewritten every
sprint stops being trusted because nobody can tell what actually changed.
So each document has a **trigger condition**, and if the condition is not met,
the document is left alone and that is the correct outcome.
**Diagrams are the exception, and only diagrams.** `docs/diagrams/*.mmd` carry
an anchor sha, so whether one is current is a fact rather than a judgement —
`/crew:diagram` refreshes those when `diagramsStale` or `diagramsMissing` fires
(see `crew-diagrams`). Nothing in the table below works that way. Whether a
change deserves a CHANGELOG entry depends on whether a user can observe it, and
no sha answers that, which is why these stay manual.
| Document | Update when | Never |
|---|---|---|
| `CHANGELOG.md` | Behaviour users or callers can observe changed | Refactors, formatting, internal renames |
| `README.md` | Setup, commands, or the mental model changed | Every ticket |
| `SECURITY.md` | Reporting process, supported versions, or a disclosed issue changed | Routine security fixes |
| `TODO.md` | Something deliberately deferred, with a reason | As a substitute for tickets |
| `docs/adr/` | A decision was made with a rejected alternative | Implementation detail |
| `docs/diagrams/` | The structure a diagram shows moved (see `crew-diagrams`); then `crew_diagrams.py embed` rewrites the README embeds | Cosmetic changes; never hand-edit a `crew-diagrams` section |
| `docs/runbooks/` | An operational procedure changed, or a new one was needed | Anything `make deploy` already does |
`/crew:implement` asks this question once per ticket. The honest answer is
usually "none of them."
## The decision is recorded, then checked
`/crew:docs <id>` writes `.work/tickets/<id>/docs.json` once: `{"reasons": {"<document>":
"<why not>"}, "deferred": [{"key", "why", "unblock"}]}`. `crew_docs_check.py --ticket <id>`
then prints each document `updated`, `not needed (<reason>)` or `MISSING`. A changed
marketplace entry's CHANGELOG line (its name and new version, under `[Unreleased]`) is
mechanical and no reason waives it; README and SECURITY.md are `not needed` only with a
recorded reason once a trigger fired; every deferral must reach TODO.md. ADRs and
runbooks stay judgement, reported `not measured`. It runs before review, never after.
## Generated blocks are not yours to edit
Some documentation is compiled from source. Editing the output is work that gets
destroyed on the next build, silently.
Before editing any markdown, check for these and stop if you are inside one:
```
<!-- BEGIN_TF_DOCS --> ... <!-- END_TF_DOCS --> terraform-docs
<!-- begin-openapi --> ... <!-- end-openapi --> openapi generators
<!-- AUTO-GENERATED --> various
```
For terraform-docs specifically, the content comes from the `/** */` header in
`main.tf`, from `footer.md`, and from the `description` fields on variables and
outputs. Change those. See `crew-terraform`.
If a README has such a block, hand-written prose belongs **above** it.
## CHANGELOG
Keep a Changelog format, semantic versions, newest first.
```markdown
## [Unreleased]
### Added
- Scheduled report Lambda replacing the sbmovement .NET job (ACME-1058)
### Fixed
- Header validation rejected files with a UTF-8 BOM
```
Write from the **ticket and the diff**, not from the conversation. One line per
user-visible change, in the language of someone using the thing rather than
someone who built it: "rejects files with a BOM" not "added BOM strip in
`validate_header()`".
Reference the ticket id. Do not paste commit hashes — they are in git already and
they make the file unreadable.
## SECURITY.md
Rarely changes, and that is fine. It should carry: how to report a vulnerability
and to whom, which versions get fixes, and expected response time.
**Never** log specific vulnerabilities here while they are unfixed. A public
`SECURITY.md` describing an open hole is a disclosure, not documentation. Fixed
issues go in the CHANGELOG once released.
## TODO.md
Only for work deliberately deferred with a reason, which is different from work
not yet started — that belongs in tickets.
```markdown
- [ ] Replace the polling loop with EventBridge — deferred until the SFTP
migration lands, because the current trigger depends on the share path
```
Each entry says **why it is deferred and what would unblock it**. Without that,
it is an inbox that grows forever and gets deleted in frustration two years later.
`.work/SMOKE-GAPS.md` is the same idea for test coverage and stays separate.
## ADRs
One decision per file, append-only, never edited after acceptance. A superseded
decision gets a new ADR that references the old one — the record of having
changed your mind is the valuable part.
Write them when the alternative was real. "We used the standard library" is not
an ADR. "We chose stored procedures over application logic because the batch
window requires set-based operations" is.
## Doing the work
```
/crew:docs # check what this change should touch, and do it
/crew:docs --audit # report staleness across all docs, change nothingFrom mbadali25/useful-claude-add-ons, GPL-2.0.
More from mbadali25/useful-claude-add-ons
- 0.603score
crew-graph
mbadali25
Build and query a code graph of this repository with graphify, keeping it on the filesystem as graph.json and fresh on commit.
SkillAFreshness: A★ 2Claude Code +9 - 0.603score
crew-best-practices
mbadali25
Community best practices for working with Claude Code - context management, planning, TDD, hooks, skills, subagents, slash commands and MCP - with what crew…
SkillAFreshness: A★ 2Claude Code +9 - 0.603score
crew-qa-standards
mbadali25
Harness, QA-review and environment standards plus a report-only audit (parallel tests, pinned linters CI really runs, gate rules that can fail, QA environment…
SkillAFreshness: A★ 2Claude Code +9 - 0.599score
crew-diagrams
mbadali25
Author architecture, process, sequence, ER and data-flow diagrams as Mermaid that pass a measured readability check (no crossing lines, nothing drawn through…
SkillAFreshness: A★ 2Claude Code +9 - 0.598score
crew-context
mbadali25
Manage context exhaustion - warn near the limit, write handoff notes, and resume work after a clear or compact.
SkillAFreshness: A★ 2Claude Code +9 - 0.598score
crew-debugging
mbadali25
Find the cause of a defect before proposing a fix - read errors, reproduce, check recent changes, gather evidence from this repo's code map and verification…
SkillAFreshness: A★ 2Claude Code +9 - 0.598score
crew-cloud
mbadali25
Configure the official AWS and Azure MCP servers safely for a repo.
SkillAFreshness: A★ 2Claude Code +9 - 0.598score
crew-memory
mbadali25
Conventions for durable project memory across repos, including Obsidian vault integration.
SkillAFreshness: A★ 2Claude Code +9 - 0.598score
crew-notify
mbadali25
Set up outbound notifications to Microsoft Teams or Telegram - deploy results and "Claude stopped and is waiting" questions - and explain the two-way MCP…
SkillAFreshness: A★ 2Claude Code +9 - 0.598score
crew-change
mbadali25
The change-request template, the ten questions a change board requires, and how each field maps onto ServiceDesk Plus, Jira and a local file.
SkillAFreshness: A★ 2Claude Code +9 - 0.598score
crew-providers
mbadali25
Set up, verify, and invoke external model providers - Codex, Kimi Code or GitHub Copilot for QA review, Gemini for design second opinions.
SkillAFreshness: A★ 2Claude Code +9 - 0.595score
crew-execute
mbadali25
Execute an approved plan task by task, in this session - TDD per step, a ruling instead of a silent deviation, and only four conditions that stop for a…
SkillAFreshness: A★ 2Claude Code +9
Signals
- Stars
- 2
- Forks
- 0
- Open issues
- 17
- Last commit (this folder)
- Oct 7, 2026 (21 days ago)
- Last release
- Jul 28, 2026 (3 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- GPL-2.0
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from mbadali25
- 0.603score
crew-qa-standards
mbadali25
Harness, QA-review and environment standards plus a report-only audit (parallel tests, pinned linters CI really runs, gate rules that can fail, QA environment…
SkillAFreshness: A★ 2Claude Code +9 - 0.603score
crew-graph
mbadali25
Build and query a code graph of this repository with graphify, keeping it on the filesystem as graph.json and fresh on commit.
SkillAFreshness: A★ 2Claude Code +9 - 0.603score
crew-best-practices
mbadali25
Community best practices for working with Claude Code - context management, planning, TDD, hooks, skills, subagents, slash commands and MCP - with what crew…
SkillAFreshness: A★ 2Claude Code +9 - 0.599score
crew-diagrams
mbadali25
Author architecture, process, sequence, ER and data-flow diagrams as Mermaid that pass a measured readability check (no crossing lines, nothing drawn through…
SkillAFreshness: A★ 2Claude Code +9 - 0.598score
crew-debugging
mbadali25
Find the cause of a defect before proposing a fix - read errors, reproduce, check recent changes, gather evidence from this repo's code map and verification…
SkillAFreshness: A★ 2Claude Code +9 - 0.598score
crew-context
mbadali25
Manage context exhaustion - warn near the limit, write handoff notes, and resume work after a clear or compact.
SkillAFreshness: A★ 2Claude Code +9
Cite this page
crew-docs skill. Agent Mart. https://agentmart.dev/skills/mbadali25-crew-docs (last updated 2026-10-24)