GitHub Repo Briefing
open-code-review
A code-review system that combines deterministic rules with an LLM agent and reports findings at exact source lines, including rules for common security and correctness issues.
A first test
Run it on a test pull request with one known defect. Compare the line comment with a human review and inspect the configured rules before connecting the service to a live repository.
claude-code
An agentic coding tool for terminal, IDE, and GitHub workflows that can inspect code, explain it, and perform routine tasks.
A first test
Give it 1 read-only question in a test repository. Check its file access and answer before authorizing edits.
ECC
A collection of engineering skills, memory, security checks, research-first workflows, and performance guidance for Claude Code, Codex, OpenCode, and Cursor.
A first test
Install one workflow in a disposable repository and run it on a small issue. Check the generated plan, tests, and permission requests before adding the full collection.
WeKnora
An open-source knowledge platform for turning documents into a searchable RAG system, reasoning agent, and wiki.
A first test
Import a few public documents and ask 1 source-specific question. Confirm the answer cites the correct passage.
gods-eye-view
A browser-based 3D globe that displays public aircraft, ship, satellite, earthquake, traffic, and camera feeds.
A first test
Open a public region in the local demo. Check the data source and timestamp before using any scene in a visual report.
agent-skills
A pack of 24 workflow skills for definition, planning, implementation, testing, review, performance, simplification, and release work in coding agents.
A first test
Install the pack with npx skills add addyosmani/agent-skills. In one real repo, run /spec, /plan, and /test on a small change, then keep it if the outputs provide useful evidence.
knowledge-work-plugins
anthropics/knowledge-work-plugins
A collection of plugins for knowledge-work tasks in Claude Cowork, with commands and workflows for roles such as research, writing, and analysis.
A first test
Install one role-specific plugin and run it on a public source. Check the generated files and tool permissions, then remove commands that do not fit your actual workflow.
i-have-adhd
A coding-agent output skill that asks for concise answers, clear next actions, and less buried information when a task has many details.
A first test
Install it and run the same small task with and without the skill. Compare the location of the answer, risks, and next action, then keep only useful changes in your project instructions.
context-mode
An MCP server that sandboxes tool output and stores searchable session events to reduce coding-agent context use.
A first test
Run the same read-only repository question with and without it. Compare context usage and check the stored session data.
worktrunk
A CLI for managing Git worktrees, with commands and hooks designed for parallel agent workflows.
A first test
Create 2 worktrees in a disposable repository. Confirm each branch and directory stays isolated before using it on a client project.
humanizer
A Markdown agent skill that rewrites AI-sounding prose while aiming to preserve its meaning.
A first test
Apply it to a draft paragraph, then compare claims line by line with the source before using the copy in a client report.
core
The open-source Home Assistant platform for local home automation, with integrations for devices and services.
A first test
Use its demo or a test instance to configure 1 automation. Check the trigger, action, and local data handling before adding devices.