The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the GitHub CLI (gh) to interact with repository metadata (milestones, labels) and to publish the final PRD and task slices as issues.
[EXTERNAL_DOWNLOADS]: The skill performs research on external websites and library documentation to provide technical context for feature development.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes data from the local codebase and external web research, which could contain malicious instructions designed to influence the generated PRD or issues.
Ingestion points: Local codebase files and external web content (SKILL.md, Step 2 and 7).
Boundary markers: No explicit delimiters or safety markers are used in the templates or instructions to separate ingested data from the agent's logic.
Capability inventory: The skill is capable of reading files, performing web research, and writing to GitHub repositories via the gh tool (SKILL.md).
Sanitization: There is no evidence of sanitization or validation of the content gathered from the codebase or research before it is included in the documentation.

snap-spec