secondbrain-discussion
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- No Executable Code (SAFE): The skill consists entirely of Markdown instructions and templates. No Python, JavaScript, or shell scripts are included or referenced for execution.
- Indirect Prompt Injection (SAFE): While the skill ingests untrusted data from user conversations to populate meeting notes, it lacks the capabilities to exploit this surface. | Ingestion points: Step 1 collects participant names, topics, and discussion points from conversation context. | Boundary markers: Data is placed into standard Markdown templates without specific delimiters. | Capability inventory: The skill only involves writing static Markdown files to local directories (docs/discussions/ and .claude/data/discussions/); no system commands or network requests are performed. | Sanitization: No explicit sanitization is required for this documentation-only use case.
- Data Exposure (SAFE): The skill accesses local project directories intended for documentation and schema storage. No sensitive system paths, credentials, or private keys are targeted.
Audit Metadata