A workflow for keeping product documentation aligned with recently merged code changes. Documentation explains how people use the product, while merged changes are code updates already accepted into the project.

Use it for scheduled or on-demand documentation audits, API documentation regeneration, validation, builds, and preparing a documentation pull request.

It finds outdated pages and public interface changes before users rely on incorrect instructions.

update-docs — Skill by stainless-code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add skills/stainless-code/codemap/update-docs

Any agent

npx skills add stainless-code/codemap --skill update-docs

Clone the repo

git clone --depth 1 https://github.com/stainless-code/codemap

Made for: Claude Code, Codex.

Per session 72 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,133 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.00072	$0.01133
Opus 5	$0.00036	$0.00566
Sonnet 5	$0.00014	$0.00227
Haiku 4.5	$0.00007	$0.00113

Measured yesterday against content hash 61f68d360acf, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

update-docs scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.agents/skills/update-docs/SKILL.md · 37 lines

How it starts

The opening of the file, as written. The whole thing — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Update docs — sync apps/docs with merged PRs

Upstream workflow: npx skills add haydenbleasel/blume --skill blume-update-docs. This local skill encodes it for scheduled agents in this repo.

When to use

Scheduled — weekly (or similar) run to catch docs drift from merged PRs.
On demand — after a batch of merged PRs that may have changed public API surfaces, recipes, schema, or Blume config.

Workflow

Collect merged PRs in the lookback window (default 7 days): git log --since="7 days ago" --merges --first-parent and gh pr list --state merged --search "merged:>$DATE". Done: every merged PR in the window listed with title, number, and surfaces touched.
Drop feature-flagged work — skip PRs/commits touching flag-gated code; never document unreleased surfaces. Done: flag-gated PRs removed from the list.
Map changed surfaces to docs pages — trace diffs in src/cli/, src/application/, src/adapters/, src/db.ts (createTables() / SCHEMA_VERSION), package.json exports, blume.config.ts / apps/docs/**, templates/recipes/*.{sql,md}, templates/agent-content/**, and TypeDoc-facing JSDoc to the pages that document them (content/reference/api/** generated, content/reference/** hand, content/guides/**, content/recipes/**, content/concepts/**, plus homepage _home/** when behavior framing changed). Regenerate API MDX via bun run docs:api when public exports/JSDoc drifted. Done: every changed public surface mapped to zero or more docs pages.
Edit only drifted facts — signatures, option tables, recipe ids, schema columns, behavior descriptions, entry-point peers. Use docs-voice; do not rewrite stable prose. Peer framing stays code-index / agent-context tools only — never store/adapter middlewares. Done: every drifted fact updated; no stable prose touched.
Verify — bun run docs:validate -- --strict && bun run docs:check -- --isolated && bun run docs:build && bun run docs:audit (run from apps/docs/ or via root docs:* scripts); fix until green. Do not commit if any fail. Per verify-after-each-step. Done: all four green.
Open or update a PR on branch docs/sync-<date> with a summary of what drifted and the fixes. Label it docs so merge deploys /codemap via FTP (docs-governance/LIFECYCLE.md § README surfaces). Follow commit/PR conventions; never --no-verify (no-bypass-hooks). Changeset for a docs-only sync is patch (pre-1.0; schema-breaking is the only minor trigger). Done: PR open with docs label and drift summary, or step 7 applies.
Clean no-op — if nothing drifted, report the PRs checked and surfaces audited; do not open a PR. Done: report lists every PR checked and every surface audited.

Read the full file on GitHub · 37 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

yesterday First seen · 37 lines · 72 tokens per session scan A 61f68d360acf

Subscribe to this mod's changes

update-docs is a skill published in the GitHub repository stainless-code/codemap (8 stars, last pushed 6d ago), licensed MIT. It adds 72 tokens to every session and 1,133 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Backend and APIs Code review Writing and docs ai-agents ai-coding ast code-graph code-index code-intelligence

roam

Codebase comprehension via roam-code CLI. Use when exploring codebases, planning modifications, debugging failures, assessing PR risk, or checking architecture health. Triggers on: understanding project structure, pre-change safety checks, finding symbols/files, blast radius analysis, affected tests, health scoring…

Cranot/roam-code · 86 tokens

analyze-usage

Use when asked to analyze codescout tool usage, check for error patterns, audit tool health, generate a usage report, spot anti-patterns, or clear usage statistics to start fresh.

mareurs/codescout · 42 tokens

m1nd-first

Use when investigating a repository, searching for implementation, reviewing changes, working from specs/docs, or preparing a risky code change in an environment where m1nd is available. This doctrine makes m1nd the first investigative layer before grep, glob, or manual file reads, except when the task is pure…

maxkle1nz/m1nd · 78 tokens

update-docs