The Agent Skills Directory

[SAFE]: The skill defines a professional workflow for vulnerability research and code analysis. The instructions focus on verifying source-to-sink paths and exploitability, which are standard security practices. No malicious patterns, hidden instructions, or dangerous commands were detected.
[NO_CODE]: The skill does not contain any executable scripts, binary files, or automated tool configurations. It operates as a set of instructions for the AI agent without requiring any local or remote execution environments.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze external files and context, creating an attack surface where the source code being reviewed could contain malicious instructions.
Ingestion points: The skill accepts code_path and recon_context as inputs in SKILL.md.
Boundary markers: The instructions do not define specific delimiters to isolate untrusted code from the agent's instructions.
Capability inventory: The skill does not define any tools, network operations, or file-write capabilities, limiting the potential impact of an injection.
Sanitization: No sanitization or escaping mechanisms are specified for the incoming data, which is typical for a code review persona.

code-review-analyst