The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted data from GitHub (issue titles, PR descriptions) which could contain malicious instructions designed to influence the agent's analysis or proposed actions.
Ingestion points: Data is fetched via gh issue list and gh pr list in SKILL.md.
Boundary markers: No explicit delimiters or warnings to ignore embedded instructions are present when processing fetched content.
Capability inventory: The skill can execute destructive actions including gh issue close, gh pr merge, gh pr close, git push origin --delete, and gh label delete as defined in Step 4 of SKILL.md.
Sanitization: No sanitization or validation of the fetched issue/PR content is specified before it is analyzed by the agent.
[COMMAND_EXECUTION]: The skill executes shell commands using the gh and git CLI tools to perform repository management.
Evidence: Step 4 of the workflow in SKILL.md lists commands for closing issues, merging/closing PRs, and deleting branches/labels.
Mitigation: The skill follows a "Human-in-the-loop" pattern, using AskUserQuestion in Step 3 to require explicit user confirmation before executing any destructive commands. It also includes a rule prohibiting deletion of protected branches.

lovstudio-gh-tidy