The Agent Skills Directory

[SAFE]: The skill is composed entirely of markdown instructions defining a process for peer review. It contains no scripts, binaries, or network-enabled commands.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it instructions the agent to read external project files.
Ingestion points: Reads COLLABORATION_PROTOCOLS.md and project-specific review scopes.
Boundary markers: No explicit boundary markers or delimiters for ingested content are specified.
Capability inventory: The skill specifies writing review verdicts to local directories under 00_project/reviews/.
Sanitization: No sanitization or validation logic is defined for the external files.

crew-review