The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes maintenance scripts (doctor.mjs, status.mjs, instructions.mjs) from the project's local .specforge/core/scripts/ directory. These operations are restricted to a defined command list and are used for project health monitoring.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface by processing untrusted project intake files while having the capability to modify files and execute local scripts.
Ingestion points: The agent reads requirement data from 00-intake/original-request.md, 00-intake/brief.md, and 00-intake/brainstorm.md.
Boundary markers: No delimiters or isolation instructions are employed when reading these files.
Capability inventory: The agent can execute Node.js scripts and update local project files like prd.md.
Sanitization: The skill does not implement validation or sanitization for the content of processed intake files.

sf-prd