Genblaze is an open source Python SDK for orchestrating generative AI media pipelines across video, audio, and image providers with built in provenance for every output.
Skill Claude CodeCodex
Prepare a wave-level genblaze release — scope version bumps and dependency-floor drift with tools/preparerelease.py, apply them, gate with make pre-release, and open a release PR. Cannot tag or publish. Use before cutting a new CHANGELOG wave (as opposed to /release-check, which gates one already-decided package…
Skill Claude CodeCodex
Pre-release gate for one genblaze package — version bump, tests, lint, typecheck, CHANGELOG entry, entry-point sanity. Use before creating a GitHub Release (which triggers the PyPI publish workflow in .github/workflows/release.yml).
Skill Claude CodeCodex
Scaffold a new genblaze provider connector package with all required files, tests, and entry points by analyzing existing connectors for conventions.
Skill Claude CodeCodex
Run tests for a single genblaze package or for only packages whose files changed since main. Use when iterating on a specific connector or after a focused edit — much faster than make test across all 13 packages.
Skill Claude CodeCodex
Audit docs/ and root markdown for staleness. Checks lastverified headers against git mtime, validates Python code examples parse, checks cross-reference links resolve. Use monthly, before release, or after any large refactor.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: