The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the repository being analyzed, which could contain malicious instructions designed to influence the agent's actions during infrastructure optimization.
Ingestion points: The scripts/discover_infra.py script reads content from package.json and scans the filesystem for various infrastructure-related file paths and directories.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to disregard potential commands embedded within the discovered file names or package.json metadata.
Capability inventory: The skill allows the agent to modify the filesystem by filling out checklists, generating reports, and implementing infrastructure changes if requested.
Sanitization: The discovery script performs basic JSON parsing of package.json but does not sanitize the resulting strings before they are presented to the agent's context.

infra-optimization