The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python scripts (resolve_customization.py) and standard library modules (webbrowser) to manage project settings and display generated reports.
[DATA_EXPOSURE]: The workflow involves reading project-specific configuration files (config.yaml) and metadata to personalize the facilitator's responses and output formatting.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes external design documents (GDDs) and user-supplied PRD drafts, which serves as a potential surface for indirect prompt injection.
Ingestion points: Document extracts in SKILL.md and references/validate.md where user inputs or existing artifacts are analyzed.
Boundary markers: The skill utilizes subagents for extraction tasks, although explicit delimiters for untrusted content are not detailed in the provided instructions.
Capability inventory: The skill has capabilities to write files (prd.md, decision-log.md, validation-report.html) and execute local Python commands.
Sanitization: No specific sanitization logic is described for the content extracted from external documents.

gds-prd