observing-log
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from user sources such as voice memos, scribbled notes, and external observing lists. This creates an attack surface for indirect prompt injection.
- Ingestion points: User-provided notes and memos entered during or after an observing session (documented in
SKILL.md). - Boundary markers: There are no explicit instructions to use delimiters (e.g., XML tags) or to ignore instructions that might be embedded within the user's notes.
- Capability inventory: The skill instructions suggest the agent can call external 'notes tools' or a 'Scratchpad connector' to save the output.
- Sanitization: The skill lacks requirements for sanitizing or escaping the user-provided text before formatting it for output or tool usage.
Audit Metadata