competition-runtime-routing

Pass

Audited by Gen Agent Trust Hub on Mar 31, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface. * Ingestion points: The skill instructs the agent to process untrusted external data from HTTP headers (Host, X-Forwarded-Host, X-Forwarded-Proto, X-Forwarded-Port, Origin), path prefixes, cookies, and live request traces (SKILL.md, references/runtime-routing.md). * Boundary markers: Absent. The skill provides no instructions to use delimiters or to treat the analyzed network data as potentially malicious content. * Capability inventory: No dangerous tool capabilities such as file-write, network operations, or subprocess calls are present in the provided files. * Sanitization: Absent. The skill does not mention escaping, validating, or filtering external inputs before they are interpolated into the prompt or processed by the agent.
  • [NO_CODE]: The skill consists entirely of instructional Markdown and YAML configuration files. No executable scripts, binaries, or command lines are present in the provided files.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 31, 2026, 12:16 PM