audit-reference-originality
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python helper script,
scripts/build_evidence_inventory.py, which performs file system operations and executes Git commands viasubprocess.check_output. The implementation uses argument lists rather than shell strings, which is a secure practice that prevents shell injection. - [INDIRECT_PROMPT_INJECTION]: The skill's primary purpose is to ingest and analyze external data, including website text, metadata, and reference materials. This creates a surface for indirect prompt injection where malicious instructions embedded in the audited content could attempt to influence the agent's audit results or behavior.
- Ingestion points: The inventory script reads text content and metadata from user-specified directories and reference files. The agent is further instructed to inspect rendered text, hidden labels, and metadata in
SKILL.md. - Boundary markers: The instructions do not define specific boundary markers or provide explicit warnings to the agent to ignore instructions contained within the ingested data.
- Capability inventory: The skill has the capability to read local files, execute Git commands (via the inventory script), and propose file modifications (fixes) based on its analysis.
- Sanitization: There is no evidence of sanitization, filtering, or escaping of the external content before it is processed by the agent or included in the final report.
Audit Metadata