The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes user-generated conversation content to populate a predefined GitHub issue template. This creates a surface where malicious instructions or formatting within the conversation could attempt to manipulate the resulting issue.
Ingestion points: Previous conversation history containing the brainstorming session as described in SKILL.md.
Boundary markers: The skill employs a structured Markdown template with explicit section headers (Problem, Proposed outcome, etc.) to organize the output.
Capability inventory: The skill utilizes the gh CLI for creating GitHub issues and labels, and writes to local .scratch/ directories.
Sanitization: A mandatory user review and confirmation step is required before the issue is created, providing a human-in-the-loop defense against potential injection.
[COMMAND_EXECUTION]: The skill uses the GitHub CLI (gh) to perform repository actions.
Evidence: Includes instructions to run gh auth status, gh label create, and gh issue create in shell environments.
Context: These operations are the core functionality of the skill and are performed only after explicit user confirmation of the drafted content.

brainstorm-to-issue