The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from GitHub issues, pull requests, and labels, which presents a surface for indirect prompt injection.
Ingestion points: Data is ingested through gh issue list, gh pr list, and gh label list commands as specified in SKILL.md.
Boundary markers: No specific boundary markers or instructions are defined to separate external data from the system prompt or to instruct the agent to ignore embedded instructions in the fetched data.
Capability inventory: The skill possesses significant repository modification capabilities, including closing issues, merging PRs, and deleting branches/labels (gh issue close, gh pr merge, git push origin --delete, gh label delete).
Sanitization: There is no evidence of content sanitization or validation for the data retrieved from GitHub before it is summarized for the user.
[COMMAND_EXECUTION]: The skill performs destructive repository operations.
Evidence: The workflow includes commands for deleting remote branches via git push origin --delete and removing labels with gh label delete.
Context: These actions are core to the skill's purpose and are gated by a mandatory AskUserQuestion step, requiring explicit user approval before execution.

lov-gh-tidy