The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands to interact with the local environment and the GitHub API.
Evidence: It invokes git remote get-url and gh auth status to check configuration.
Evidence: It uses gh issue view, gh label create, gh issue create, and gh api to read PRDs and publish plan sub-issues.
Context: These operations are limited to repository metadata and issue tracking, which are central to the skill's purpose of project planning.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external Product Requirement Documents, which represents a potential attack surface for indirect prompt injection.
Ingestion points: PRD content is loaded from local Markdown files, GitHub issue bodies, or the current conversation context.
Boundary markers: The instructions do not specify the use of delimiters (e.g., XML tags or backticks) when passing the PRD content to the planning agent.
Capability inventory: The skill has the capability to write files to the local repository and create/edit issues on GitHub.
Sanitization: While filenames and titles are slugified, the PRD body content is not explicitly sanitized. However, the skill mandates a 'Verify the plan' stage (Step 6) that includes an independent review (via sub-agent) and final human approval before execution, significantly mitigating the risk of executing malicious instructions contained within a PRD.

prd-to-plan