A collection of deck.gl map layers for autonomy visualization

Autonomy-first opencode suite — ships features end-to-end without babysitting. Agents, commands, and opinionated config for long-horizon coding.

Compile an agent profile onto a substrate (GitHub Actions or a local termfleet loop) — the open-autonomy CLI.

Guarded autonomy state machine, risk classifier, and delivery orchestration for osstack.

Autonomy IRL

Lightweight functional utility library complementing ES6

CLI for MoltNet — AI agent identity and autonomy network

Local AI coding-agent usage dashboard — tokens, cost, projects, autonomy.

Governed autonomy for any repo. An off-by-default autonomy loop that adopts your repo's rules, proposes small reviewable changes, and gates them in CI so weakened tests cannot merge.

MELRA autonomy kernel and command-line interface.

Merge autonomy for AI-authored code. Turn your repo's standards into evidence-backed readiness reports the agent reads back.

Baseline Autonomy — Autonomous AI Operations Layer

pi extension package for autonomy-control workflows in pi

MELRA autonomy kernel with MCP and loopback HTTP interfaces.

Trust engine with 5-level autonomy, escalation, demotion, and audit trail

HALO core - loop state machine, port resolution, preflight, autonomy checks

Interaction modes for Reactive Agents — 5 autonomy modes, checkpoints, and human collaboration

Eval-gated autonomy: deterministic rules + LLM rubric + an N-clean-runs gate that decides when an AI agent has earned unattended runs.

TypeScript client for the Vaara HTTP API: accountable autonomy for every autonomous action. Conformal risk scoring, policy gating, hash-chained tamper-evident audit, named detectors.

Earned-autonomy governance for AI agents — agents earn autonomy the way employees earn trust.