The Agent Skills Directory

[SAFE]: No security issues were detected. The skill provides legitimate documentation and workflow guidelines for an AI agent to use the grepai command-line tool for semantic code exploration.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function involves searching and reading external codebase content, which inherently presents a surface for indirect prompt injection. This is a standard risk for any code exploration tool and is not indicative of malicious intent within the skill itself.
Ingestion points: Code previews and search results returned by the grepai search command and call graph relationships from grepai trace.
Boundary markers: None specified in the instructions for delimiting search output.
Capability inventory: The skill utilizes grepai for search/trace operations and the Read tool to examine files.
Sanitization: No specific sanitization or filtering of ingested codebase content is implemented or mentioned.

mcp-grepai