The Agent Skills Directory

Indirect Prompt Injection (LOW): The skill processes external PRD files which could contain malicious instructions designed to hijack the agent's behavior.
Ingestion points: The skill reads local files and their references as part of the implementation workflow.
Boundary markers: No delimiters or warnings to ignore embedded instructions are used in the prompt logic.
Capability inventory: The skill uses powerful tools including Write, Edit, and Bash.
Sanitization: No input validation or sanitization is performed on the data read from files.
Command Execution (SAFE): While the Bash tool is enabled, it is primarily used for grep searches to ensure codebase consistency. No malicious command strings or obfuscated shells were found.

prd-implementation-precheck