The Agent Skills Directory

[PROMPT_INJECTION]: The skill instructs the agent to analyze external, untrusted code changes, which creates a surface for indirect prompt injection attacks where malicious instructions could be embedded in the reviewed code.
Ingestion points: PR descriptions and code diffs (SKILL.md).
Boundary markers: Absent; no delimiters or instructions are provided to the agent to treat external content as data only.
Capability inventory: None; the skill contains no executable scripts (SKILL.md).
Sanitization: Absent; the checklist does not include validation or sanitization of the input code prior to analysis.
[EXTERNAL_DOWNLOADS]: The documentation includes an installation command npx clawhub@latest which references an external package registry and a tool that is not part of the trusted vendor or service lists.
[COMMAND_EXECUTION]: The skill documentation suggests the use of the npx command for installation, which involves executing code from an external source on the host system.
[NO_CODE]: The skill consists entirely of instructional Markdown content and does not include any executable scripts, binary files, or configuration-level code, which limits direct execution risks.

code-review