Skip to content
Agentmart
Submit a tool

wiki

Use when generating, updating, auditing, or checking the status of a codebase Markdown wiki (setup, outline, sync, status, audit, stamp), or when asked what…

Skill by AJpon

IndexedMIT
Write a review

Chalk Labs

Need help deploying wiki?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

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/ajpon-wiki

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.

We use your email only for this newsletter. Privacy

About

Use when generating, updating, auditing, or checking the status of a codebase Markdown wiki (setup, outline, sync, status, audit, stamp), or when asked what to do next with the wiki or how agentic-doc works.

Use cases: documents, security

No 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 and code. Checked Nov 1, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found
Runs as
Python package

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/wiki

What's inside

Skill name
wiki
Description
Use when generating, updating, auditing, or checking the status of a codebase Markdown wiki (setup, outline, sync, status, audit, stamp), or when asked what to do next with the wiki or how agentic-doc works.
License
MIT
Instructions
92 lines in SKILL.md · view on GitHub

Before installing

  • Ships executable scripts

Files (62)

  • SKILL.md
  • .gitignore.in · assets/work/
  • audit-drift.md · references/
  • common.md · references/
  • outline-areas.md · references/
  • outline-merge.md · references/
  • outline-propose.md · references/
  • outline-survey.md · references/
  • overview.md · references/
  • run-audit.md · references/
  • run-outline.md · references/
  • run-sync.md · references/
  • step-command.md · references/
  • sync-fix.md · references/
  • sync-integrate.md · references/
  • sync-research.md · references/
  • sync-write.md · references/
  • verify-citation.md · references/
  • verify-claim.md · references/
  • writing-rules.md · references/
  • catalogue.schema.json · schemas/
  • command-result.schema.json · schemas/
  • config.schema.json · schemas/
  • drift.schema.json · schemas/
  • findings.schema.json · schemas/
  • fix.schema.json · schemas/
  • integrate.schema.json · schemas/
  • outline-result.schema.json · schemas/
  • proposal.schema.json · schemas/
  • research.schema.json · schemas/
  • survey-areas.schema.json · schemas/
  • survey.schema.json · schemas/
  • verdict.schema.json · schemas/
  • write.schema.json · schemas/
  • __init__.py · scripts/adoc/
  • cache.py · scripts/adoc/
  • catalogue.py · scripts/adoc/
  • cli.py · scripts/adoc/
  • cmd_cache.py · scripts/adoc/
  • cmd_catalogue.py · scripts/adoc/
  • cmd_checkpoint.py · scripts/adoc/
  • cmd_config.py · scripts/adoc/
  • cmd_findings.py · scripts/adoc/
  • cmd_plan.py · scripts/adoc/
  • cmd_post.py · scripts/adoc/
  • cmd_report.py · scripts/adoc/
  • cmd_stamp.py · scripts/adoc/
  • cmd_status.py · scripts/adoc/
  • cmd_toc.py · scripts/adoc/
  • cmd_unverify.py · scripts/adoc/
  • commits.py · scripts/adoc/
  • config.py · scripts/adoc/
  • dates.py · scripts/adoc/
  • errors.py · scripts/adoc/
  • gitutil.py · scripts/adoc/
  • i18n.py · scripts/adoc/
  • lint.py · scripts/adoc/
  • mermaid_check.mjs · scripts/adoc/
  • state.py · scripts/adoc/
  • textutil.py · scripts/adoc/
  • …and 2 more

SKILL.md (first lines)

# agentic-doc wiki

This skill generates a Markdown wiki from a codebase and keeps it in step with the code. The user runs it as `/agentic-doc:wiki <mode> [arguments]`. The modes are `help`, `setup`, `outline`, `sync`, `status`, `audit` and `stamp`.

## Paths

`<SKILL_DIR>` is the directory that contains this `SKILL.md`. In the commands below, replace it with that absolute path. Run every command from the repository root:

```bash
uv run --script "<SKILL_DIR>/scripts/agentic_doc.py" status
```

Keep the double quotes around the script path. Exit codes: 0 means OK, 1 means findings were reported, 2 means the input was refused or the command failed. Read `<SKILL_DIR>/references/overview.md` for the modes, the flow and the page states.

## Checks before each mode

Skip these checks for `help` and `setup`.

1. Run `uv run --script "<SKILL_DIR>/scripts/agentic_doc.py" config check`.
   - If it exits with 1 or 2, show its output, explain that the configuration must be fixed or created first, point to `setup`, and stop.
2. Run `uv run --script "<SKILL_DIR>/scripts/agentic_doc.py" config show`.
   - Except in `outline`: if `catalogue_exists` is false, point to `outline` and stop.
   - In `outline`: if `catalogue_exists` is true and the user did not pass `--force`, stop. Explain that `outline` would replace the catalogue, and that every page would become unverified (`source_commit` is reset). Point to `outline --force` if the user really wants that.

## Modes

| Mode | What to do |
| --- | --- |
| No mode, `help`, or a question | 1. Run `config check`. If it passes and a catalogue exists, run `status`.<br>2. Read `<SKILL_DIR>/references/overview.md`.<br>3. Tell the user where the project is in the flow, and the next command to run, with its arguments.<br>4. For questions about the scripts, run `uv run --script "<SKILL_DIR>/scripts/agentic_doc.py" --help`. |
| `setup` | Use the `wiki-setup` skill. If you cannot invoke skills, read `<SKILL_DIR>/../wiki-setup/SKILL.md` and follow it. |
| `status [<id>...] [--state <state>] [--verbose]` | Run `uv run --script "<SKILL_DIR>/scripts/agentic_doc.py" status` with the user's arguments, and show its output. |
| `stamp [--sha <rev>] [--pages <id>...] [--dry-run]` | Run `uv run --script "<SKILL_DIR>/scripts/agentic_doc.py" stamp` with the user's arguments, and show its output. Exit code 1 means some pages were skipped. Explain why, using the output. |
| `outline [--force]`, `sync [--pages <id>...] [--include-unresolved]`, `audit [--lint-only]` | See the next section. |

## Running outline, sync and audit

### When the Workflow tool is available (Claude Code)

Run the workflow `agentic-doc:wiki-run` with these args:

```json
{"mode": "<outline|sync|audit>", "plugin_root": "${CLAUDE_PLUGIN_ROOT}"}
```

Add the user's options to the args:

| Option | Arg |
| --- | --- |
| `--pages a b` | `"pages": ["a", "b"]` |
| `--include-unresolved` | `"include_unresolved": true` |
| `--force` | `"force": true` |
| `--lint-only` | `"lint_only": true` |

When the workflow returns, report the result:

- **sync**:
  1. Show a table with one row per page: `id`, `result` and `findings`.
  2. List `failed`, and anything in `integrate.notes` or `integrate.broken_links_remaining`.
  3. Run `status`, and give the next step from its `Next:` line.
- **audit**:
  1. Write `{"results": <drift>, "failed": <drift_failed>}` as JSON to `.agentic-doc/work/drift.json` with your file-writing tool. Do not pipe it through the shell.
  2. Run the command below, and show its output. With `--lint-only`, run `report` without `--drift`.

  ```bash
  uv run --script "<SKILL_DIR>/scripts/agentic_doc.py" report --drift .agentic-doc/work/drift.json
  ```

- **outline**:
  - If `installed` is true, give the catalogue path and the page count. Ask the user to review and edit the catalogue (chapters, pages, `sources`, `brief`) before the first `sync`.
  - Otherwise, show `issues`.

If the workflow fails with an error, show the error. Do not run the steps by hand instead.

### When there is no Workflow tool (GitHub Copilot, Codex and others)

Run the same steps yourself, one at a time. Read the procedure for the mode and follow it:

- `outline`: `<SKILL_DIR>/references/run-outline.md`
- `sync`: `<SKILL_DIR>/references/run-sync.md`
- `audit`: `<SKILL_DIR>/references/run-audit.md`

## What you may change

- Write only:
  - The wiki directory.
  - The catalogue.
  - `.agentic-doc/`.
  - `.agentic-doc.config.json` (in `setup`).
  - Files that a `post_commands` hook changed and that the integrate step restores.
- Do not commit. Leave the commit to the user, and remind them that a wiki page should be committed together with the sources it describes.

From AJpon/agentic-doc, MIT.

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Oct 23, 2026 (14 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from AJpon

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get wiki running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying wiki?

CallGet help
Cite this page

wiki skill. Agent Mart. https://agentmart.dev/skills/ajpon-wiki (last updated 2026-10-24)