The Agent Skills Directory

[SAFE]: The skill is purely instructional, providing the agent with a methodology for planning and dependency tracing. It does not invoke external scripts, download remote content, or request elevated privileges.
[INDIRECT_PROMPT_INJECTION]: The skill defines a workflow for analyzing user requests and repository code to derive dependency graphs. While this involves processing untrusted external data (repository files and user input), the skill does not grant the agent additional dangerous capabilities that could be exploited via injection beyond its standard operational scope.
[SAFE]: The metadata, description, and frontmatter are consistent with the skill's stated purpose of planning and design.

dependency-trace