The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local scripts and tools to prepare and verify releases.
Evidence: Runs python3 scripts/plugin-release.py preflight <plugin> <version>, python3 -m unittest discover -s scripts -p 'test_plugin_release.py' -v, and actionlint (SKILL.md).
[EXTERNAL_DOWNLOADS]: The skill interacts with well-known package registries to publish and verify artifacts.
Evidence: Mentions distribution and verification against npm, PyPI, and ClawHub registries (SKILL.md).
[INDIRECT_PROMPT_INJECTION]: The skill has a potential surface for indirect prompt injection due to processing external files while possessing significant capabilities.
Ingestion points: Reads and processes integrations.json, package manifests, RELEASING.md, and CHANGELOG.md (SKILL.md).
Boundary markers: No specific boundary markers or instructions to ignore embedded commands within these external files are specified.
Capability inventory: The skill can execute local scripts, interact with external registries, and modify repository files such as manifests and lockfiles (SKILL.md).
Sanitization: The skill includes verification steps such as actionlint and artifact integrity checks, but does not explicitly define sanitization for natural language instructions in external data.

plugin-release