#ai-coding-agent
VinvAI
VinvAI
An autonomous swarm that finds bugs, dead code and performance issues in your codebase — and fixes them with closed-loop RL. Nine oracles hunt a real run, your own coding agent does the edit, and nothing lands unless it passes acceptance tests the agent never sees. Local, no API keys.
sync.md
sync-md
Keeps AI coding-agent instruction files (AGENTS.md, CLAUDE.md, .cursorrules, .github/copilot-instructions.md) in sync by using an AI model to compare their meaning, not just their text. Also includes local-only length and active-file checks. No telemetry, no bundled API keys.
Yuhi Code — See What Your AI Agent Sees
yuhi-ai-labs
Prepare your workspace before Claude Code sees it. One Privacy Mode across Static Prepare, Dynamic Terminal, and Native Claude GUI — transform direct personal identifiers, redact secrets, and keep the session isolated from your normal VS Code environment.
AgentHours: AI Client Time Tracker
CavalryCollective
Automatic billable time tracking for the AI era — hands-on coding plus Claude Code and Codex CLI runs, token usage, and AI cost by client project. Accurate timesheets with no timers to start or stop.