search-decisions

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill retrieves decision content and rationale from the Jupi platform, which could contain adversarial instructions.
  • Ingestion points: Output from search-decisions-tool and get-decision tools as described in SKILL.md.
  • Boundary markers: Absent; the skill does not use delimiters to isolate retrieved data from agent instructions.
  • Capability inventory: The skill utilizes tools to read decision data; this external input could potentially influence other tools available to the agent.
  • Sanitization: No content filtering or validation is performed on the data fetched from the external source.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 04:48 PM
Security Audit — agent-trust-hub — search-decisions