sf-close
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local Node.js scripts located in the
.specforge/core/scripts/directory, such asarchive-work.mjsanddoctor.mjs, to automate project lifecycle management and archiving. - [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it reads and processes data from multiple local development artifacts to generate release and rollback documentation.
- Ingestion points: Reads various project files including
work.yaml,report.md, andtechnical-design.md. - Boundary markers: No explicit delimiters or boundary markers are used to isolate untrusted data from the instruction context.
- Capability inventory: The skill executes shell commands via Node.js for project administration and reporting.
- Sanitization: Ingested data is used to generate reports without documented sanitization or filtering steps.
- [SAFE]: The skill implements safety measures by explicitly refusing to install or synchronize agent skills to external platforms without explicit user consent.
Audit Metadata