The Agent Skills Directory

[SAFE]: The skill provides documentation and usage guidelines for a legitimate code analysis utility. The operations described are restricted to local repository inspection and do not involve credential exposure, persistence, or unauthorized network activity.
[INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external source code data via the calldiff tool, creating a surface for potential indirect prompt injection if the analyzed code contains malicious instructions. However, this is inherent to the tool's primary purpose of code analysis.
Ingestion points: The tool reads from local source files and git history during execution.
Boundary markers: The instructions do not define specific delimiters for the output of the CLI commands.
Capability inventory: The skill uses calldiff exclusively for structural observation, such as finding call paths, structural diffs, and module outlines.
Sanitization: The instructions do not include steps to sanitize or validate the tool output before agent processing.

calldiff