pentest-race-conditions

Pass

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION] (LOW): Indirect Prompt Injection Surface. The skill is designed to ingest and process data from external target applications using tools with significant capabilities.
  • Ingestion points: The agent processes untrusted data from target HTTP responses (status codes and bodies) via tools like aiohttp_race, httpx_parallel, requests_threaded, and state_checker. It also processes data from direct database query results via the db_query tool.
  • Boundary markers: Absent. There are no delimiters or instructions provided in references/workflows.md or elsewhere to ensure the agent ignores potentially malicious commands embedded in the data it retrieves from targets.
  • Capability inventory: Across its scripts and references, the skill provides tools for microsecond-synchronized parallel HTTP requests and the ability to execute arbitrary SQL queries on databases.
  • Sanitization: Absent. No sanitization, escaping, or validation logic is defined for the external content before it is processed by the agent or used in subsequent analysis steps.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 18, 2026, 05:57 PM