The Agent Skills Directory

[SAFE]: The skill establishes a robust and disciplined archaeological process for codebase analysis. It includes mandatory quality gates and requires the agent to maintain a clear distinction between neutral documentation and critical assessment, effectively preventing unauthorized or unsafe operations.
[COMMAND_EXECUTION]: The skill instructions direct the agent to use standard local shell utilities, such as find, grep, mkdir, and cloc, to discover repository structures and gather file metrics. These commands are essential for the skill's primary function and are used safely within the context of architectural discovery.
[INDIRECT_PROMPT_INJECTION]: By design, this skill ingests and processes untrusted source code from external codebases. This creates a surface for indirect prompt injection. The skill mitigates this risk by employing a delegation model where subagents perform detailed code reading and return only summarized findings to the coordinator, thereby protecting the core context from potential malicious instructions embedded in the analyzed code.

using-system-archaeologist