deepresearch-integrator
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill processes external research files which could contain instructions designed to influence the agent's output or tool usage. 1. Ingestion point: Step 2.1 in SKILL.md reads source files. 2. Boundary markers: Absent. No specific instructions are provided to the agent to disregard instructions within the data. 3. Capability inventory: allowed-tools: "*" allows for broad tool access including file manipulation. 4. Sanitization: Absent. Content is summarized and integrated without filtering potential injection content.
- Data Exposure & Exfiltration (SAFE): No hardcoded credentials or unauthorized data exfiltration patterns were detected.
- Remote Code Execution (SAFE): No patterns of remote script downloading or dynamic code execution were found in the skill content.
Audit Metadata