The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands to retrieve git configuration and repository history.
Evidence: git config user.name, git log main..HEAD --oneline, git diff main..HEAD --stat in SKILL.md.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources which could contain malicious instructions.
Ingestion points: Commit messages, Linear ticket descriptions, and repository source code (SKILL.md).
Boundary markers: Absent. The instructions do not specify how to distinguish between data and instructions during processing.
Capability inventory: Shell command execution (git), file system read access, and file system write access (RFC.md).
Sanitization: Absent. No mention of filtering or escaping external content.

rfc-writer