template-repo
View on GitHubAgent orchestration & security template featuring MCP tool building, agent2agent workflows, mechanistic interpretability on sleeper agents, and agent integration via CLI wrappers
Reference template for orchestrating multiple AI coding agents (Claude, OpenCode, Crush) via a GitHub Projects task board, approval-gated claims, 20+ containerized MCP servers and AI PR review. Also ships research packages for sleeper-agent detection and agent trust/security hardening.
Use Cases
Run a council of AI coding agents on one shared repoDelegate tasks with a GitHub Projects v2 board work queueGate agent work behind explicit human approval commentsIntegrate 20+ containerized MCP servers for code quality, content, 3D, video and speechAutomate AI PR review and agent iteration on feedbackBuild hardened multi-stage CI for agent-authored codeDetect sleeper-agent behavior via residual stream analysis and linear probesSimulate autonomous economic agentsIntegrate agents into legacy software via injection, shared memory IPC and overlay renderingGenerate images, video and speech through agent-driven pipelines
Built With
- Language
- Rust
- Frameworks
- Model Context Protocol (MCP) · Claude Code · Docker · Docker Compose · GitHub Actions · OpenRouter · ComfyUI · Blender · ElevenLabs · pytest · Ruff · Terraform · Cargo · LaTeX · GitHub Projects v2
Tags
agent-orchestration · multi-agent · mcp · mcp-servers · a2a · agent-security · ai-safety · mechanistic-interpretability · sleeper-agents · agent-governance · claude-code · docker · ci-cd · github-actions · code-review · text2image