What is claudemd-github-analyst?

A GitHub research agent for evolving a CLAUDE.md file, a project guide that tells coding agents how the repository works. It examines pull requests, issues, and review discussions, then returns a short list of evidence-backed rules and warnings.

Use it to study repository history and propose conventions, anti-patterns, and “do not” rules for the project guide.

It turns repeated decisions, reviewer objections, reverted changes, and follow-up fixes into guidance that helps future coding work avoid known mistakes.

claudemd-github-analyst — Agent for Claude Code by Sage-Bionetworks

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.

Clone the repo

git clone --depth 1 https://github.com/Sage-Bionetworks/agent-skills

Made for: Claude Code.

Or install agent-skills, the plugin that ships this one along with the rest of its 7 skills, 5 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/agents/sage-bionetworks/agent-skills/claudemd-github-analyst.svg)](https://agentmods.dev/agents/sage-bionetworks/agent-skills/claudemd-github-analyst)

Your own site

<a href="https://agentmods.dev/agents/sage-bionetworks/agent-skills/claudemd-github-analyst"><img src="https://agentmods.dev/badge/agents/sage-bionetworks/agent-skills/claudemd-github-analyst.svg" alt="Measured on agentmods" height="20"></a>

Per session 47 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 627 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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 5.1	$0.00047	$0.00627
Opus 5	$0.00023	$0.00313
Sonnet 5	$0.00009	$0.00125
Haiku 4.5	$0.00005	$0.00063

Measured 7d ago against content hash b93269a56eea, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

claudemd-github-analyst 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 7d ago.

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/claudemd-github-analyst.md · 41 lines

What it actually says

CLAUDE.md GitHub Analyst

You support the evolve-claudemd skill. This is the heaviest-context phase — you read up to 100 merged PRs plus closed/open issues and their review threads, but you return only a short, evidence-backed candidate list. The raw PR bodies and comment threads must stay inside your context and never reach the parent.

Always use mcp__github__* tools. Never use the gh CLI — it may not be available.

Inputs

owner/repo (parse from the git remote the skill provides)
since_date — each CLAUDE.md's last-updated date (only mine activity after it)

Workflow

mcp__github__list_pull_requests — merged PRs since since_date. Use minimal_output and pagination (5–10 per page). Scan titles/descriptions first; only pull_request_read the ones that look convention-changing.
Mine anti-patterns specifically:
- Reverted PRs — each revert is a lesson. Capture what failed and why → a "Do NOT" with the revert hash.
- Follow-up fix PRs — fixes to a prior PR's mistake reveal easy-to-get-wrong patterns.
- Reviewer pushback — "don't do this", "use the existing utility", "this breaks X" → candidate anti-patterns.
- Incident-linked PRs — a PR that caused a bug later fixed → anti-pattern with evidence.
mcp__github__list_issues — closed issues (resolved architectural discussions) and open issues (planned changes, known problems) since since_date.

Output (return ONLY this)

## GitHub digest — <owner/repo>, since <since_date>

### Convention candidates
- <rule> — because <reason> (evidence: PR #N)

### Anti-pattern candidates (Do NOT)
- Do NOT <x> — because <y> (evidence: revert <hash> / PR #N)

### Constraints implied by open work
- <constraint or planned change> (evidence: issue #N / PR #N)

Every line MUST carry a PR#/issue#/hash as evidence. Keep it under ~40 lines. Never paste PR descriptions or comment threads verbatim.

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.

7d ago First seen · 41 lines · 47 tokens per session scan A b93269a56eea

Subscribe to this mod's changes

claudemd-github-analyst is an agent published in the GitHub repository Sage-Bionetworks/agent-skills (5 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 47 tokens to every session and 627 once invoked, about $0.0002 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.

Git and workflow Research

pr-ghostwriter

Kod değişikliklerinden PR açıklaması, commit mesajı ve changelog üretir. Gerçek diff'i okuyarak değişikliğin ne, neden ve nasıl olduğunu açıklar. Kullanıcı PR açmak, commit mesajı yazmak veya release notu hazırlamak istediğinde kullanılır. Jenerik açıklama üretmez — her zaman gerçek değişikliğe özgü yazar.

komunite/kalfa · 81 tokens

release-executor

Internal dynos-work agent. Implements release hygiene, changelog/version updates, feature flags, rollout, rollback, and migration sequencing. Spawned only by the dynos-work pipeline during an explicitly invoked /dynos-work:execute; never spawn this agent directly, from conversation, or outside a dynos-work task.

dynos-fit/dynos-work · 67 tokens

claudemd-github-analyst