The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection by processing untrusted data.
Ingestion points: User-provided {feature} argument and PRD files read from the docs/00-pm/ directory.
Boundary markers: None present in the skill instructions to delimit untrusted content.
Capability inventory: The skill has access to Read, Write, Edit, Glob, Grep, and Bash tools as defined in the SKILL.md frontmatter.
Sanitization: There is no evidence of input validation or content sanitization for the feature names or ingested documents.
[DATA_EXFILTRATION]: The skill is configured to read and write product documentation within the project's subdirectories. No patterns suggesting unauthorized access to sensitive system files or credentials were detected.
[EXTERNAL_DOWNLOADS]: The skill provides an informational link to a GitHub repository for methodological credit. This is a reference to a well-known service and does not trigger remote code execution.

pm-discovery