The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch and analyze content from external GitHub repositories using the gh CLI and git clone. This introduces a vulnerability surface where malicious instructions embedded in repository files (e.g., README.md, source code comments, or metadata) could be processed by the agent, potentially leading to unintended actions.
Ingestion points: Remote repository content retrieved via gh api and git clone operations (SKILL.md, REFERENCES.md).
Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded prompts within the processed data.
Capability inventory: The skill possesses the ability to execute shell commands (gh, git), access the local file system (Read, Glob, Grep), and perform network operations.
Sanitization: There is no evidence of sanitization or filtering of the fetched content before it is analyzed by the agent.
[SAFE]: The skill references well-known technology organizations and services for its examples and installation guidance (e.g., Facebook, Vercel, Homebrew, and official Linux package managers). These references are legitimate and do not pose a security risk.
[SAFE]: The skill includes explicit safety instructions requiring user confirmation before executing state-changing or destructive operations, such as merging pull requests or deleting repositories.

github-navigator