Guide · Agent skill
How to use the GitHub Author Context agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The github-author-context skill is a set of instructions and helper steps that tells an AI agent how to build a short background summary on the person who opened a pull request. It is written for maintainers who review PRs and want quick context on contributors they do not know. After reading this guide you will know what the skill does, how to install it, how to ask for it in plain language, and what it needs to run safely.
Quick start
- 1Download and unzip. Download the skill ZIP from this page and unzip it into ~/.claude/skills/ or a project's .claude/skills/ folder.
- 2Or upload in the Claude apps. Open Settings, go to Capabilities, then Skills, and upload the skill.
- 3Install and log in to gh. Make sure the GitHub CLI, git, and ripgrep are installed and that gh is authenticated.
- 4Ask during a PR review. Request author context for a specific PR number in plain language.
- 5Check the output. Review the author context block the agent adds near the top of the review and verify the facts against the GitHub profile.
- 6Record notes sparingly. Only add a contributor note when it will help future reviews, using the maintainer helper script.
What is an agent skill and what does this one do?
An agent skill is a folder containing a SKILL.md file with instructions that an AI agent can load when a task matches the skill's description. This skill covers GitHub contributor context: identity, activity, trust, and company or team signals.
When it is active, the agent gathers public and local information about the PR author. That includes their GitHub profile, recent merged and open pull requests and issues in the target repository, and local git history. It then writes a short 'author context' block near the top of a code review.
- Who: name, company, location, or role, with a confidence level
- Activity: merged and open PRs, issues, and reviews or commits if known
- OpenClaw signal: maintainer, candidate, drive-by, vendor, security, or unknown
- Risk: review load, broad PRs, low history, company governance, or none obvious
Install the github-author-context skill
Download the ZIP for this skill from this page and unzip it. For a personal setup, place the unzipped folder in ~/.claude/skills/. For a single project, place it in that project's .claude/skills/ folder.
You can also upload the skill in the Claude apps. Open Settings, go to Capabilities, then Skills, and upload the skill there.
The package includes two files: agents/openai.yaml and SKILL.md.
- Personal install: ~/.claude/skills/
- Project install: .claude/skills/ inside your project
- Claude apps: Settings → Capabilities → Skills
Install the required tools
The skill calls command-line tools. It needs the GitHub CLI (gh), installed and authenticated, plus git and ripgrep (rg). Without gh authenticated, the GitHub lookups will not work.
The skill also refers to local maintainer scripts at ~/Projects/maintainers/scripts/clawtributors and contributor folders. These are not included in this skill, so the source-order steps that depend on them will not work until you have them.
- GitHub CLI (gh), installed and logged in
- git
- ripgrep (rg)
- Local clawtributors script and contributor folders (not included)
gh pr view <n> --json author,url,headRepository,baseRepository -q '{author:.author.login,url:.url,repo:.baseRepository.nameWithOwner}'How to trigger the skill with natural requests
You do not need a special command. Ask the agent in plain language during a PR review. The skill is meant to be used by default during PR review, so a request that mentions a PR or an author is usually enough.
The skill says to skip the profile pass for the user steipete unless you explicitly ask for it.
- 'Review PR 123 and give me background on the author.'
- 'Who is the author of this pull request? Check their GitHub activity in this repo.'
- 'Add author context near the top of the review for PR 456.'
- 'Look up the GitHub profile and recent merged PRs for this contributor.'
Understand the lookup steps the agent follows
The agent works in a fixed order. It first checks local OpenClaw contributor notes if they exist. Then it reads the live public GitHub profile using gh api. Next it counts activity in the target repository using gh search for merged PRs, open PRs, and open issues, and checks the collaborator permission. Finally it looks at local git history for the author.
The skill asks the agent to read only specific sections of contributor files, such as Identity, Signals, Context, Evidence, and Notes, rather than whole files.
- Step 1: local contributor notes (clawtributors find github <login>)
- Step 2: live GitHub profile (gh api users/<login>)
- Step 3: target-repo activity (gh search prs and gh search issues)
- Step 4: local git evidence (git log and git shortlog)
gh api "users/<login>" --jq '{login,name,company,location,bio,blog,twitter_username,created_at,followers,following,public_repos}'Use the author context block and contributor notes
The output is a short block placed near the top of a PR review. Keep it brief. The skill also explains when to record a note after a merge, rejection, close, or review. A note is added only when it will help future reviews, such as a first good merge, repeated quality problems, or identity confirmation. Ordinary activity is not recorded.
Notes are written with the maintainer repo helper so the Markdown stays consistent. The skill says not to quote private phone or email details unless the user asks for them.
- Keep notes terse, factual, dated, and linked
- Do not record ordinary noise
- Separate an author's employer from company-directed OpenClaw work
~/Projects/maintainers/scripts/clawtributors note github <login> --kind merged --pr <n> --summary "Focused bug fix. Tests credible. Smooth review."What people use it for
- Adding a short background block to a review of a PR from an unfamiliar contributor
- Checking whether an author has prior merged PRs or open issues in the same repository
- Spotting review-load or broad-PR risks before reading a large diff
- Recording a terse note after a first good merge or a repeated quality problem
- Adapting the lookup pattern to your own contributor-tracking scripts
Frequently asked questions
Is the github-author-context skill free?
The skill is published under the MIT licence, as stated on this page. Check the licence text in the source repository before reusing it in your own project.
What do I need to install before using it?
You need the GitHub CLI (gh) installed and authenticated, plus git and ripgrep (rg). The skill also refers to local maintainer scripts and contributor folders that are not included in the download.
How do I install the github-author-context skill?
Download the ZIP from this page and unzip it into ~/.claude/skills/ for personal use or into .claude/skills/ inside a project. You can also upload it in the Claude apps under Settings, then Capabilities, then Skills.
Why does the skill not work on my machine?
The most likely cause is that gh is not installed or not logged in, or that the ~/Projects/maintainers/scripts/clawtributors script is missing. The skill depends on these paths and on OpenClaw-specific conventions, so it will not run as-is elsewhere.
Does the skill run scripts and use the network?
Yes. The skill runs scripts and makes network calls to GitHub through gh. It also reads local git history and may write notes to maintainer helper files, so review what it runs before using it.
Is this a general-purpose GitHub profile tool?
No. It is built for the OpenClaw maintainer workflow and expects specific paths. If you want a general profile tool, you would need to adapt the steps. It should also not be used on people without consent or for hiring or harassment-style judgments, since it compiles employer, location, and activity data.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
This skill suits maintainers who review pull requests and want quick background on unfamiliar contributors, especially if they follow the OpenClaw maintainer workflow or are ready to adapt its steps. It is a poor fit for anyone wanting a general profile tool, or for hiring or harassment-style use, because it depends on local paths and gathers personal data about individuals.
Download the skill