AgentPack
Local context engine that prepares task-focused file packs for AI coding agents
Visit vishal2612200.github.io ↗What is AgentPack?
AgentPack is a local context engine designed to help AI coding agents work more effectively on software projects. It analyzes repositories to identify and rank relevant files, then builds compact context packs tailored to specific tasks. The tool integrates with Claude Code, Cursor, Codex, Windsurf, and other AI agents through plugins and MCP tools, supporting both interactive workflows and CI environments.
The system tracks project state, workflow progress, and validation evidence across sessions without requiring hosted indexing. Users can run simple commands to describe changes, finish work, and learn from completed tasks. AgentPack was built to work locally and offline, making it suitable for teams that need verifiable, benchmarkable file selection without external dependencies.
Momentum
Daily activity over the last 30 days: upvotes, unique visits, new GitHub stars and Hacker News points.
How do we know it's vibe coded?
vibetrend discovered AgentPack automatically. Evidence: GitHub vishal2612200/agentpack; repo has .claude, .codex, .cursor. View source.
Similar vibe-coded apps


Prescriptive methodology and agent skills for spec-driven development with AI


Shared persistent memory for AI coding tools across sessions and applications

In-browser behavior tree execution engine for AI agents at ~60Hz

Local plugin that makes AI coding agents provide evidence and resist flattery


Browser and desktop 2.5D + 3D CAD/CAM workspace for CNC machining