doc-brd-fixer

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local system utilities like sha256sum and grep to perform hash validation and drift detection. It also invokes a project-specific script, validate_brd_wrapper.sh, for structural compliance checks.\n- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it processes external data to drive document modifications.\n
  • Ingestion points: The skill reads from audit reports (BRD-NN.A_audit_report_vNNN.md) and upstream documents to identify required fixes and determine if a document should be archived or regenerated.\n
  • Boundary markers: No explicit delimiters or boundary markers are used to isolate ingested data from the skill's internal logic, creating a risk that malicious data in a report could influence agent behavior.\n
  • Capability inventory: The skill is capable of creating, moving, and writing to files on the local filesystem, as well as executing a local shell-based validation wrapper.\n
  • Sanitization: There is no evidence of sanitization or content validation for the data extracted from audit reports before it is used to modify documents or transition between fix phases.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 11:05 AM