Ante

A ghost in your shell. Ante is a self-contained coding agent that lives in your terminal and self-organizes. One ~15MB compressed download from Antigma Labs that expands to a single Rust executable with zero runtime dependencies, built to get the most out of any model.

It works like Claude Code or Codex, with none of their dependencies or model constraints. It can also be the optimized core for building your own harness and high-performing assistants.

The three are one design decision. An agent you can verify, afford, and run anywhere is light enough to run by the thousands: the substrate for self-organizing intelligence. See Philosophy.

Beyond the headline numbers​

• Zero vendor lock-in. Bring your own API key, subscription, or local model. Switch between 12+ providers freely — Anthropic, OpenAI, Gemini, Grok, Open Router, and more. No account required, no telemetry gate, no walled garden. Not even ours.
• Multi-agent orchestration. Coordinate multiple agents that self-organize around complex tasks.
• Extensible. Custom skills, sub-agents, MCP, and persistent memory across sessions.
• Obsessed with simplicity and quality. In a world of abundance, maintaining quality and a high level of trust is more valuable than ever.

How it works​

Ante runs as a client-daemon architecture. The daemon manages structured turn lifecycles — prompt, tool calls, confirmation, execution — channeling model power through a safe, predictable loop. The client can be an interactive TUI, a headless CLI, an external process, or a channel bot via ante gateway.

Architecturally, Ante is moving toward an agent-centric runtime where long-lived agents coordinate through message passing — an actor model with hierarchical supervision, isolated state, and clear ownership of resources per agent.

Ante can also run as a long-lived server (ante serve) that external clients drive through a structured JSONL protocol — ideal for building editor plugins, web UIs, and custom integrations.

Next steps​