The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes content from the repository, including CONTEXT.md, package.json, and .git/config, to determine scaffolding requirements. This creates a potential surface where malicious instructions embedded in project files could influence the agent during setup.
Ingestion points: Reads git remote -v, CONTEXT.md, and package.json as specified in SKILL.md.
Boundary markers: No explicit boundaries or instructions to ignore embedded commands are present in the exploration logic.
Capability inventory: Includes file system writes to docs/agents/ and executing CLI commands through git, gh, and glab tools.
Sanitization: No validation or sanitization is performed on the ingested file content.
[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using git to inspect repository configuration and uses the GitHub (gh) and GitLab (glab) CLI tools to interact with issue trackers and pull/merge requests as part of its core functionality.

setup-skills