operational-risk-narratives
Pass
Audited by Gen Agent Trust Hub on Mar 3, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists exclusively of Markdown documentation and templates for risk analysis; it contains no executable scripts, binaries, or configuration files that could be run on a system.
- [PROMPT_INJECTION]: The skill defines a framework for processing external data such as incident reports, which represents an indirect prompt injection surface. This is assessed as safe due to the lack of tools or system-level permissions.
- [PROMPT_INJECTION]: Mandatory Evidence Chain for Indirect Injection: (1) Ingestion points: User-provided event details, financial impact, and control environment data as defined in the methodology of SKILL.md. (2) Boundary markers: Absent; the templates do not specify delimiters or instructions to ignore embedded commands in external data. (3) Capability inventory: None; the skill defines no tool calls, subprocesses, file operations, or network requests. (4) Sanitization: Absent; the skill does not prescribe validation or filtering of input content.
Audit Metadata