GitHub Repo Briefing
ai-agent-book
An open Chinese book about AI-agent design and engineering, with the full text, a compiled PDF, and code examples arranged by chapter.
A first test
Read one chapter and run its matching example. Change one input, compare the output with the explanation, and record the dependency versions before using the pattern elsewhere.
worldmonitor
A web dashboard that combines news aggregation, geopolitical monitoring, maps, and infrastructure signals into a real-time situational view.
A first test
Run npm install and npm run dev with test settings. Open one event and trace it to the source link and timestamp before using the dashboard for an operational decision.
code-review-graph
A local code-intelligence graph for MCP and CLI clients that maps a repository so agents can focus on relevant files during code review.
A first test
Index a small repository and ask for one review path. Compare the returned files with the actual call graph, then check the context reduction before trusting a large-repo review.
jcode
A RAM-efficient coding-agent harness for running an agent from the terminal with a small local footprint.
A first test
Install it with the project script and point it at a disposable clone. Ask for one test-only change, measure memory use, and inspect the diff before any commit step.
pi-web
A web interface for the pi coding agent that lets users start and observe coding-agent sessions from a browser.
A first test
Run it against a disposable repository and complete one read-only task. Check login, terminal output, working directory, and session cleanup before exposing the web interface.
OmniRoute
A local AI gateway that exposes one endpoint for many model providers and translates requests for clients such as Claude Code, Codex, Cursor, and Copilot.
A first test
Run it on the local port with one provider and one low-risk prompt. Inspect the selected model, request log, costs, and fallback behavior before adding credentials for other providers.
kimi-code
A Kimi Code command-line agent for software development tasks in a terminal workspace.
A first test
Install it in a disposable clone and ask it to explain one failing test. Inspect the model request, shell commands, and proposed patch before allowing a write.
pi
An AI-agent toolkit with a unified LLM API, agent loop, terminal UI, and coding-agent CLI for building or running local agent workflows.
A first test
Run the CLI with one configured model and a read-only repository task. Save the tool trace, check the API endpoint, and test cancellation before using it in automation.
kimi-cli
A command-line coding agent from Moonshot AI for working with repositories and developer tasks in a terminal session.
A first test
Install it in a test repository and ask for a file explanation. Verify the command list, model configuration, and final output before granting write or network access.
ai-engineering-from-scratch
rohitg00/ai-engineering-from-scratch
A hands-on course that teaches AI engineering by building and shipping small projects with an agent as a tutor.
A first test
Install its skills and run /start-learning for one topic. Make the smallest exercise yourself, run its test, and record what you changed before asking the agent for the next lesson.
skills
A collection of small engineering skills for planning, implementation, testing, review, simplification, and release work. The files are editable and can be installed as a managed Claude Code plugin.
A first test
Install one skill set, not both installation methods. Run its planning and testing commands on a small change, then keep only the files that improve your actual review evidence.
RuView
A WiFi-sensing system that estimates presence, movement, and some vital-sign signals from radio data without using a camera.
A first test
Run ruview doctor and ruview verify in a test room. Compare the output with a known movement test and treat any health-related signal as experimental, not a medical measurement.