repokeeper 0.1.0

AI-powered open source maintainer agent

RepoKeeper 🤖

AI-powered open source maintainer agent.

RepoKeeper automates the tedious parts of open source maintenance: community monitoring, dependency updates, CI diagnosis, stale issue detection, and AI-powered PR creation — all driven by a single YAML profile that describes your maintainer preferences.

Modules

#	Module	What it does
🔭	Community Radar	Monitors GitHub issues/discussions for keywords, AI-classifies posts, drafts issues, sends notifications
🔍	Daily Patrol	Scans dependencies, diagnoses CI failures, finds stale issues, generates health scores
🤖	Implementation Agent	Reads issues + codebase, implements code changes, opens a PR for your review
👤	Maintainer Profile	YAML config describing your style, tone, PR standards, and tech preferences

Quick Start

Install the CLI:

pip install repokeeper

Create a maintainer profile:

repokeeper init --workflows

# repokeeper.yml — place this in any repo
maintainer: your-github-username

radar:
  keywords: [bug, crash, security, feature request]

patrol:
  schedule: "0 8 * * 1-5"

agent:
  model: deepseek-chat

Then label an issue agent-todo or comment @repokeeper go.

You can also run modules directly:

repokeeper profile validate
repokeeper radar --repo owner/repo
repokeeper patrol --repo owner/repo --summary
repokeeper agent --repo owner/repo --issue 42

Documentation

Full docs: https://shenxianpeng.github.io/repokeeper

• Getting Started
• Module 1: Community Radar
• Module 2: Daily Patrol
• Module 3: Implementation Agent
• Module 4: Maintainer Profile

Requirements

• Python 3.11+
• GitHub Actions enabled
• DeepSeek API key (or any OpenAI-compatible API)

For the Implementation Agent, enable Settings → Actions → General → Allow GitHub Actions to create and approve pull requests, or provide a REPOKEEPER_GITHUB_TOKEN secret with write access to contents and pull requests.

License

MIT