smithery.ai

afml-docs-loop

Build and run an AFML-grounded, chapter-by-chapter documentation loop for OpenQuant docs. Use this when you need consistent module docs, hierarchical chapter coverage, and iterative progress tracking with MCP-backed prompts.

First seen Mar 29, 2026

Installation

$ npx skills add https://smithery.ai

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from smithery.ai · top by installs.

npx skills add https://smithery.ai

Browse all from smithery.ai

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 2,582 B
  • docs SUMMARY.md 246 B

History

  1. First seen on skills.sh
  2. First recorded snapshot · 1 installs

SKILL.md

AFML Docs Loop

Use this skill when the goal is to produce complete, consistent OpenQuant docs aligned to AFML structure.

What this skill does

  1. Builds a hierarchical docs backlog (chapter -> module section) from AFML + current module docs.
  2. Generates per-section writing prompts grounded in AFML MCP retrieval anchors.
  3. Runs a loop: claim next section, write docs, review UI, mark done.

Required script

  • Main orchestrator: scripts/afmldocsloop.py
  • Wrapper: scripts/runafmldocs_loop.sh

Quick start

# 1) initialize plan + prompt files
skills/afml-docs-loop/scripts/run_afml_docs_loop.sh init

# 2) inspect progress
skills/afml-docs-loop/scripts/run_afml_docs_loop.sh status

# 3) run MCP evidence loop across sections (writes evidence.json/evidence.md)
skills/afml-docs-loop/scripts/run_afml_docs_loop.sh evidence

# 4) claim next section and print its prompt
skills/afml-docs-loop/scripts/run_afml_docs_loop.sh next --print-prompt

# 5) mark section complete
skills/afml-docs-loop/scripts/run_afml_docs_loop.sh done <section_id> --note "what was updated"

# 6) export loop status into Astro data
skills/afml-docs-loop/scripts/run_afml_docs_loop.sh export

Iterative workflow (mandatory)

For each section:

  1. Run evidence to refresh AFML retrieval artifacts for all targeted sections.
  2. Claim next section with next --print-prompt.
  3. Use AFML MCP tools (afmlsearch, afmlget_context) to gather chapter/module grounding.
  4. Update module docs in docs-site/src/data/moduleDocs.ts.
  5. Validate rendering in Astro and apply the UI rubric in references/ui-ux-rubric.md.
  6. Mark section complete with done.
  7. Repeat until status shows no pending sections.

File targets

  • Module content source: docs-site/src/data/moduleDocs.ts
  • Module renderer: docs-site/src/pages/module/[slug].astro
  • Subject index: docs-site/src/pages/modules.astro
  • State file: docs/afml-docs-loop/state.json
  • MCP evidence: docs/afml-docs-loop/evidence.json, docs/afml-docs-loop/evidence.md
  • Generated prompts: docs/afml-docs-loop/prompts/*.md

Quality bar

  • Use references/doc-writing-template.md for consistency.
  • Use references/ui-ux-rubric.md to avoid generic/AI-slop design.
  • Keep examples executable and API-accurate for the current crate.