autocontext

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. The npm CLI and TUI require Node.js 22.19.0 or newer; contributors should use the version pinned in ts/.nvmrc.

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, openrouter, claude-cli, codex, pi-rpc, or another provider when you need that runtime. See agent integration for the full matrix.

Running it on your own GPU instead? Self-hosted models covers the whole loop on vLLM, Ollama, or any OpenAI-compatible endpoint — including what each role actually resolves to, and why constrained output matters more on open weights. Self-hosted endpoints can additionally declare AUTOCONTEXT_PROVIDER_HOSTING=local and a fast, mid_tier, or frontier AUTOCONTEXT_PROVIDER_CAPABILITY; role-specific endpoints use matching <ROLE>_PROVIDER_* declarations.

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 serve mcp or bunx autoctx serve mcp 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/
└── context_bundles/{bundles,candidates,promotions,active.json}

Everything is filesystem-first: inspect it, diff it, replay it, export it, or feed it into training. Coach and architect context changes are stored as immutable candidates and are not served until matched candidate/incumbent trials confirm them. The live serving boundary can additionally require a cancellable independent audit and a durable campaign-wide false-promotion budget; exact causal credit is accepted only from verified single-component manifest additions. See context bundles and outcome-gated promotion. Controlled component trials feed ablation-backed attribution, so prompt selection can demote low-value context without presenting edit-size correlation as causal.

Core Surfaces

Running bare autoctx shows the concise paved-road workflow. Use autoctx --help --all in the npm CLI or autoctx commands --all in the Python CLI for the full catalog. --iterations is the primary iteration flag; --gens is a compatibility alias. autoctx --version --json reports the package version and runtime (python or typescript).

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 0.16.1

• A clearer paved-road CLI: run now requires an explicit scenario, --iterations is the primary spelling, concise help leads with the core workflow, and autoctx commands --all exposes the full catalog. Existing --gens and legacy command aliases remain available for compatibility.
• Stable machine-readable inspection: status, show, and watch --ndjson emit versioned, schema-backed envelopes with consistent stdout, stderr, and exit-code behavior. Status and show accept --run-id, while --version --json identifies the Python runtime and package version.
• More reliable run inspection: watch no longer exits before a completed generation becomes visible, and show defaults to the latest generation while rejecting conflicting selectors instead of guessing.
• Portable exports by default: export now writes JSON to stdout when no output path is supplied, supports Pi packages, and retains strategy as a compatibility alias for the default JSON format.
• Contracts ship with the package: CLI contract v2 schemas and shared fixtures are included in wheels and source distributions so downstream tools can validate the same status, show, queue, and export shapes as the CLI.
• Ratcheted package boundaries: domain, analytics, configuration, and storage implementations now follow enforced dependency directions while legacy module paths remain available as compatibility shims.

npm runtime highlights included in 0.16.1

The aligned autoctx@0.16.1 package also carries the TypeScript-first runtime work introduced in 0.16.0 and hardened in 0.16.1:

• Host-owned live composition: typed runtime capabilities, scoped cleanup and effect policies, reactive component graphs, and durable transactional activation/rollback for trusted hosts.
• A production-oriented operator TUI: the pi-tui client supports local and remote attachment, durable replay, run control and inspection, and bounded, redacted terminal state on Node.js 22.19+.
• Image-aware interactive sessions: compatible TypeScript providers can advertise image_attachments_v1; attachment validation is bounded and fail-closed before provider inference.
• Protocol and terminal hardening: exact capability negotiation, protected priority controls, bounded WebSocket resources, credential redaction, and terminal-control sanitization are enforced across the interactive path.

Python parity for the pi-tui client and image attachments remains deferred. See the TypeScript guide, runtime composition contracts, and the full changelog for details.

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/internal/scenario-parity-matrix.md for parity details.

Package Guides

Project Signals

Acknowledgments

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