Skip to main content

Hakowan MCP

Provider-neutral agent integration for Hakowan.

hakowan-mcp exposes Hakowan's deterministic inspection, schema, validation, compilation, rendering, observation, camera-fitting, and patch operations over MCP. Model selection, credentials, conversation state, and reasoning remain the responsibility of the MCP host.

Install

pip install hakowan-mcp

Optional observation support:

pip install 'hakowan-mcp[observe]'
playwright install chromium

Run

hakowan-mcp --root /path/to/project

Streamable HTTP:

hakowan-mcp --root /path/to/project \
  --transport streamable-http --host 127.0.0.1 --port 8000

See docs/mcp.md for tools, resources, prompts, host configuration, path confinement, and transport details.

Token-efficient workflow

The server returns compact schema catalogs, spec handles, validation results, and visual evidence by default. Agents should start with inspect_data and get_spec_template, chain later calls with spec_id, and request full schemas, specifications, or manifests only when needed.

Evaluation

The installed evaluator targets canonical Hakowan FigureSpec JSON:

hakowan-mcp-eval --provider reference
python -m hakowan_mcp.eval.mcp_harness \
  --model github-copilot/gpt-5-mini \
  --gallery ../hakowan-gallery \
  --output /tmp/hakowan.responses.json

See docs/evaluation.md. Generated responses and reports are intentionally excluded from Git; CI or release assets should retain them.

Responsibility boundary

Hakowan owns the schema and deterministic visualization behavior. This package owns MCP transport, agent instructions, gallery grounding, host execution, and LLM evaluation. It imports Hakowan's public APIs and does not fork its schema or validator.

The initial implementation was extracted from Hakowan commit 1012b00. This release targets the Hakowan 0.6 series and CI verifies compatibility with v0.6.0.

Release files for hakowan-mcp 0.1.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for hakowan-mcp 0.1.0
File Size Uploaded
hakowan_mcp-0.1.0.tar.gz 42.1 kB Details

Built distribution (wheel)

Table of built distributions (wheels) for hakowan-mcp 0.1.0
File Interpreter ABI Platform
hakowan_mcp-0.1.0-py3-none-any.whl Python 3 none any Details

Total release size: 91.0 kB

Release files / hakowan_mcp-0.1.0.tar.gz

Download URL hakowan_mcp-0.1.0.tar.gz
Size 42.1 kB
Tags Source
SHA-256 checksum
How to use checksums
77bed2789cd7c6677dedad2fa9bd4da49fc185a8ffb8ed0e51328ef46bc606d1
BLAKE2b-256 checksum
How to use checksums
ddbc499af4634a070142b2962241e701a64bbc76eb5238da91762be60c5b8c7a
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Sep 26, 2026.

Transparency log

Release files / hakowan_mcp-0.1.0-py3-none-any.whl

Download URL hakowan_mcp-0.1.0-py3-none-any.whl
Size 48.9 kB
Tags Python 3
SHA-256 checksum
How to use checksums
b7053e13b74de30e6433f89f2f5216e6d34c60a3ba82966de9847ad7da80ba70
BLAKE2b-256 checksum
How to use checksums
3a88bdade590176bba679f2084b3ad2b70353d0a51371cded34acaa4b877deb3
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Sep 26, 2026.

Transparency log

Release history Release notifications | RSS feed

This release

0.1.0 This release

2 release files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page