binlee-compliance-risk

Pass

Audited by Gen Agent Trust Hub on Jul 16, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill identifies a potential attack surface for indirect prompt injection through its document processing workflow.\n
  • Ingestion points: The skill reads and evaluates content from ../binlee-source-library/references/evidence-policy.md and related corpora.\n
  • Boundary markers: There are no instructions or delimiters defining the start and end of untrusted data to prevent the agent from executing instructions found within the files.\n
  • Capability inventory: The skill's capabilities are limited to text processing and analysis; it does not request network access or shell command execution.\n
  • Sanitization: No sanitization or filtering logic is present to validate the content of the external files being processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 16, 2026, 01:23 AM
Security Audit — agent-trust-hub — binlee-compliance-risk