The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes local repository documentation (README.md, docs/upstream.md) and existing skill files to generate new skill definitions.
Ingestion points: The agent is instructed to read README.md, docs/upstream.md, and other skill files to identify functional gaps.
Boundary markers: There are no explicit instructions to use delimiters or ignore instructions found within those files.
Capability inventory: The skill possesses the capability to write new files (skills/<name>/SKILL.md) and execute commands via pnpm for validation.
Sanitization: No explicit sanitization or validation of the ingested content is mentioned, which is common for development tools where the user controls the repository content.

extend-mvskills