The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external data such as code diffs and plan/specification documents. This creates an indirect prompt injection surface where content in the files being reviewed could contain instructions aimed at manipulating the agent's review verdict. However, the risk is minimal as the skill lacks automated tools for data exfiltration or system modification.\n
Ingestion points: Workflow section in SKILL.md specifies reading git diff and requirements documents.\n
Boundary markers: Absent; the skill does not instruct the agent to use specific delimiters when processing external content.\n
Capability inventory: The skill is purely instructional and contains no scripts, network operations, or file-writing tools.\n
Sanitization: No validation or filtering of external input is specified.\n- [NO_CODE]: The skill consists exclusively of markdown documentation and process guidelines without any accompanying executable scripts or configuration files.

review