GitHub - greyhaven-ai/autocontext: a recursive self-improving harness designed to help your agents (and future iterations of those agents) succeed on any task

a recursive self-improving harness designed to help your agents (and future iterations of those agents) succeed on any task

autocontext is a harness for agent improvement. Give it a goal, it runs the task against evaluation, keeps the useful lessons, discards dead ends, and leaves traces, reports, playbooks, datasets, and optional local-model training artifacts for the next run.

Docs: autocontext.ai/docs · quickstart · CLI reference · changelog

Install

The PyPI package is autocontext; the CLI is autoctx. The npm package is autoctx (not the unrelated autocontext npm package). Provider variables live in .env.example.

30-Second Run

Pi is the lowest-friction provider because it uses your local agent auth:

AUTOCONTEXT_AGENT_PROVIDER=pi \
AUTOCONTEXT_PI_COMMAND=pi \
autoctx solve "improve customer-support replies for billing disputes" --iterations 3

Use AUTOCONTEXT_AGENT_PROVIDER=anthropic, openai-compatible, claude-cli, codex, pi-rpc, or another provider when you need that runtime. See agent integration for the full matrix.

Agent Entry Points

• Pi: install pi-autocontext, then ask Pi to solve, judge, improve, list, or inspect runs through the packaged skill.
• MCP clients: run autoctx mcp-serve or bunx autoctx mcp-serve and expose the tools to Claude Code, Cursor, or another MCP client.
• Hermes: export the CLI-first skill with uv run autoctx hermes export-skill --with-references --json.

Full setup: autocontext/docs/agent-integration.md.

What A Run Leaves Behind

runs/<run_id>/
├── trace.jsonl
├── generations/<n>/{strategy.json,analysis.md,score.json}
├── report.md
└── artifacts/

knowledge/<scenario>/
├── playbook.md
├── hints.md
└── tools/

Everything is filesystem-first: inspect it, diff it, replay it, export it, or feed it into training.

Core Surfaces

Python owns the full control-plane package; TypeScript owns several operator-facing surfaces, the TUI, and Node runtime adapters. Start with autocontext/README.md or ts/README.md.

What's New in 0.11.0

• Guardrail parity fixes the CLI agent-task path so it applies the same guardrail-adjusted score threshold as the task runner, evaluated against the fully prepared task state.
• Branded id types add RunId, ScenarioName, and DbPath to the TypeScript package root; GenerationRunner.run now takes a RunId (construct with asRunId), a compile-time-only change.
• Dependency refresh clears every critical and high security advisory across the Python and TypeScript packages, including the anthropic sdk, urllib3, starlette, and vitest lines.

Scenario Families

The shipped families cover games, agent tasks, simulations, artifact editing, investigations, workflows, negotiation, schema evolution, tool fragility, operator loops, and coordination. Python and TypeScript share the family vocabulary; see docs/scenario-parity-matrix.md for parity details.

Package Guides

Project Signals

Acknowledgments

Thanks to George for generously donating the autocontext name on PyPI.