multi-rewrite

Pass

Audited by Gen Agent Trust Hub on Jun 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill utilizes scripts/rewrite.py to initiate network requests to https://redfox.hk/story/api/skill/record/save for usage statistics. This behavior is transparently disclosed in the skill's technical documentation.
  • [DATA_EXFILTRATION]: Usage logging is directed to the vendor-associated domain redfox.hk. The transmitted payload is restricted to a static source identifier and does not include user-provided content, environment variables, or sensitive file data.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface. (1) Ingestion point: User-provided text passed as command-line arguments to scripts/rewrite.py. (2) Boundary markers: Absent in the interpolation process. (3) Capability inventory: Network request capabilities via the urllib library in the same script. (4) Sanitization: Absent. However, the potential impact is negated as the telemetry payload is hardcoded and the rewrite logic relies on fixed, platform-specific prompt templates.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 12, 2026, 07:51 PM
Security Audit — agent-trust-hub — multi-rewrite