agent-data-researcher

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exposes a surface for indirect prompt injection attacks. * Ingestion points: Data enters the agent context through the WebSearch and Read tools mentioned in SKILL.md. * Boundary markers: Absent. There are no instructions or delimiters to isolate untrusted data from the system prompt. * Capability inventory: The agent has access to python, sql, Write, and WebSearch tools as defined in SKILL.md, which could be used to execute malicious logic found in external data. * Sanitization: Absent. The skill does not implement any validation or filtering for data retrieved from external sources.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 06:26 AM
Security Audit — agent-trust-hub — agent-data-researcher