The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the gh CLI to interact with GitHub for repository management, issue syncing, and status tracking. The implementation in scripts/coach.py utilizes subprocess.run with a list of arguments, which is a secure method that prevents shell injection vulnerabilities.
[EXTERNAL_DOWNLOADS]: Fetches repository metadata and commit history from GitHub's official services via the gh tool to provide status updates to the user.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data from GitHub, such as repository descriptions and commit messages, which represents a potential attack surface for indirect prompt injection if the linked repository is controlled by a malicious actor.
Ingestion points: Functions get_repo_info and get_repo_commits in scripts/coach.py retrieve data from GitHub via the gh CLI.
Boundary markers: None identified in the prompt instructions for isolating external repository data.
Capability inventory: The skill has the ability to write to the local file system (~/.openclaw/idea-coach/ideas.json) and perform GitHub operations (repo/issue creation) via the gh CLI.
Sanitization: Employs a slugify function to sanitize repository names, though repository descriptions and issue bodies are passed as direct arguments to the CLI tool.

idea-coach