The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute npx jscpd commands to compare folders. While this is the intended functionality of the tool, it involves running external code within the agent's environment.
[EXTERNAL_DOWNLOADS]: The skill facilitates the download of a 548 MB semantic model using the --semantic-download flag. The instructions correctly advise the agent to seek user confirmation before initiating the download.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data in the form of codebases provided by the user. Maliciously crafted code or comments within these files could attempt to influence the agent's summary or explanation of the comparison results.
Ingestion points: Code files within the folders specified for comparison (e.g., billing-py/, billing-ts/).
Boundary markers: No specific delimiters are defined to separate the tool's output or the code content from the agent's internal instructions.
Capability inventory: Shell command execution (npx), file reading, and report generation (HTML, JSON, Markdown).
Sanitization: No explicit sanitization or filtering of the compared code content is mentioned before it is processed by the agent.

compare-codebases