The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external GitHub issues, comments, and project items which can contain malicious instructions. * Ingestion points: Tools like mcp__github__issue_read, mcp__github__list_issues, and gh api GET requests in SKILL.md and references/search.md. * Boundary markers: The instructions lack specific delimiters or warnings to treat ingested issue content as untrusted data. * Capability inventory: The skill has extensive write capabilities via gh api and various MCP write tools such as mcp__github__issue_write. * Sanitization: No sanitization or validation of the fetched issue content is described in the workflow.
[COMMAND_EXECUTION]: The skill frequently uses the GitHub CLI (gh) to perform repository management and API operations. * Evidence: Numerous gh api and gh label commands are documented throughout the skill and its reference files.

github-issues