The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external code and metadata from pull requests, creating a surface for indirect prompt injection.
Ingestion points: Untrusted content from pull request titles, bodies, and diffs are fetched using git diff and gh pr view as specified in SKILL.md.
Boundary markers: The instructions do not provide explicit delimiters or instructions for the agent to isolate the fetched PR content from its own execution logic.
Capability inventory: The skill employs powerful command-line tools including git, gh, go test, make, and the dtctl vendor-specific CLI.
Sanitization: No specific sanitization or validation mechanisms are defined for the data retrieved from remote pull requests.

pr-review