Loki Mode is an autonomous AI coding agent platform that transforms tasks, issues, or PRDs into verified pull requests with signed receipts. It operates as a multi-provider orchestration engine (Claude Code, Codex, Aider, etc.) with a staged pipeline: intake, plan, wall (acceptance tests), implement, verify, fix, seal, and PR. The system produces Ed25519-signed evidence receipts that can be independently verified, ensuring honest outcomes. It includes a control plane (Bun/Hono + SQLite) for run persistence and a web dashboard, plus extensive integrations (Slack, Jira, Linear, GitHub, GitLab). The architecture is polyglot: a bash CLI (autonomy/loki) and a TypeScript/Bun rewrite (loki-ts) run in parallel with parity enforcement, plus Python services for memory, learning, and the dashboard. Key entry points are the `loki` CLI, the control plane server on port 47821, and the FastAPI dashboard. External dependencies include git, Claude Code CLI, various LLM providers, and optional services like OpenTelemetry and Helm for deployment.
CodexMCP is a lightweight MCP (Model Context Protocol) server that acts as a bridge between Claude Code and OpenAI's Codex CLI. It exposes a single `codex` tool that allows MCP clients to invoke Codex CLI sessions programmatically over stdio. The server handles session persistence via UUIDs, supports multi-turn conversations, parallel execution, and sandboxed workspace modes. The core implementation consists of a subprocess wrapper that streams JSON lines from Codex's `exec` command, parses agent messages and thread IDs, and returns structured results. It also includes Windows-specific command-line escaping to handle prompt injection safely. The codebase is minimal (two source modules plus a CLI entry point) but demonstrates solid patterns for wrapping external CLI tools in the MCP protocol, with graceful shutdown handling and error reporting. It does not include tests or CI/CD configuration, but the architecture is clean and focused.