alterlab-nmc-digital-ethics
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection due to its core function of autonomously searching the web and reading project files for analysis. Malicious instructions embedded in external web pages or user-provided documents could potentially influence the agent's behavior during the file creation and self-review phases.
- Ingestion points: Web search for frameworks, governance policies, and case studies; reading local project files (e.g., course materials, draft analyses).
- Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instruction' directives when processing untrusted data.
- Capability inventory: The agent has capabilities for web search, reading files from the project directory, and writing markdown deliverables back to the file system.
- Sanitization: There are no explicit requirements for the agent to sanitize or validate external content before incorporating it into its analytical output.
Audit Metadata