The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and process user-provided source and target codebases, as well as JSON reports generated by the jscpd tool. This creates a surface for indirect prompt injection where malicious instructions embedded in code comments or data files could influence the agent's analysis or porting suggestions. While the use of structured JSON for reporting provides some boundaries, there is no explicit sanitization step mentioned for the external content.
[COMMAND_EXECUTION]: The skill utilizes the jscpd command-line utility via npx to perform code comparisons and generate reports. This involves executing shell commands to interact with the file system and run the analysis tool on the provided paths.
[EXTERNAL_DOWNLOADS]: The instructions include a setup step that downloads a semantic model (CodeRankEmbed, approximately 548 MB) using the npx jscpd --semantic-download command. This involves fetching external binary data from the tool's distribution servers to enable semantic code comparison.

code-migration