The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data in the form of code snippets, diffs, and pull requests.
Ingestion points: The workflow in SKILL.md explicitly instructs the agent to analyze user-provided code, diffs, and pull requests.
Boundary markers: There are no explicit instructions or delimiters defined to separate the instructions from the potentially malicious content within the code being reviewed.
Capability inventory: This skill does not define its own tools or scripts; it relies on the agent's native capabilities to perform the review.
Sanitization: The instructions do not include steps to sanitize or escape the content of the code before processing it, which could allow embedded instructions in the code (e.g., in comments) to influence the agent.
[NO_CODE]: The skill consists entirely of markdown instructions and does not include any executable scripts, binaries, or configuration files that could hide malicious logic.

code-review-checklist