The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it reads and processes data from external files (manifests and phase artifacts) that could potentially contain malicious instructions.
Ingestion points: Reads from docs/reverse/{analysis}/manifest.json and artifact files associated with Phase 1, 2, and 3.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are specified in the prompt logic.
Capability inventory: The agent has access to Read, Write, Bash, Grep, and Glob tools.
Sanitization: No content sanitization or validation is performed on the ingested data before it is written to the final output file.
[COMMAND_EXECUTION]: The skill requests access to the Bash tool and interpolates the analysis argument into file paths. While this provides a potential surface for directory traversal or unintended command execution if input is not sanitized, the instructions limit usage to standard file management within the analysis directory.
[NO_CODE]: This skill consists entirely of markdown instructions and does not distribute any standalone scripts, binaries, or third-party code packages.

re-generate-report