The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external GitHub star lists and repository metadata, which are subsequently used to rebuild and commit tables to a project's README.
Ingestion points: The skill scrapes GitHub star lists and repo-specific file contents (such as AGENTS.md and CLAUDE.md) in SKILL.md.
Boundary markers: There are no explicit instructions to use delimiters or sanitization when processing these external strings before writing them to the repository files.
Capability inventory: The skill has the capability to write files, perform git commit, and git push to a remote repository as described in SKILL.md.
Sanitization: No sanitization logic for external repository names or metadata is mentioned before this data is interpolated into the README tables.

contributions-update