decision-quality-audit
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill operates by ingesting and analyzing external decision records, creating a surface for indirect prompt injection attacks where malicious instructions in the data could influence the agent's behavior.
- Ingestion points: The skill instructs the agent to gather and reference decision records and charters using the
@filesyntax inSKILL.md. - Boundary markers: There are no boundary markers or instructions to treat the external file content as untrusted data or to ignore embedded instructions.
- Capability inventory: The skill utilizes file-read operations to audit documents and file-write operations to save the results in the
audits/directory. - Sanitization: The skill lacks any sanitization or filtering logic for the content it processes from external files.
- [NO_CODE]: The provided skill contains only instructional markdown and configuration metadata. There are no scripts (e.g., Python, JavaScript) or binary executables included in the package.
Audit Metadata