rf-keyword-builder
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a legitimate development tool for Robot Framework syntax generation and does not contain any malicious code, obfuscation, or unauthorized access mechanisms.
- [PROMPT_INJECTION]: Evaluation of the indirect prompt injection surface (Category 8): Untrusted JSON data is ingested via file or stdin in
scripts/keyword_builder.py. The output artifact lacks explicit boundary markers to isolate interpolated strings. The skill has read access to the local project filesystem for style analysis. Sanitization is minimal, limited to newline escaping in documentation. The overall risk is assessed as safe given the tool's intended role as a formatter. - [COMMAND_EXECUTION]: The skill executes a local Python script to perform formatting tasks, which is a standard and safe operation.
- [DATA_EXFILTRATION]: The skill scans for
.robotand.resourcefiles in the project directory to identify style conventions but does not access sensitive system files or perform network operations.
Audit Metadata