The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses git log and ripgrep (rg) to inspect commit history and search for code patterns such as TODOs, error handling, and specific keywords. These are standard diagnostic commands used for the skill's intended purpose.
[INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface by ingesting data from external files and git history which could contain malicious instructions.
Ingestion points: Data enters the agent context through git log output and ripgrep search results across the codebase.
Boundary markers: No explicit delimiters are used to separate ingested data from agent instructions.
Capability inventory: The skill has read-only access to files and git history via standard CLI tools; it does not have write or network capabilities.
Sanitization: None; the skill focuses on summarizing findings for the user to review.

blindspot-pass