assignment-general

Pass

Audited by Gen Agent Trust Hub on Feb 21, 2026

Risk Level: SAFE
Full Analysis
  • Prompt Injection (SAFE): No evidence of instructions attempting to bypass safety filters or override system prompts. The instructions are focused on providing a structured legal analysis workflow.
  • Data Exposure & Exfiltration (SAFE): No hardcoded credentials, sensitive file paths, or network operations were detected. The skill operates entirely within the chat context.
  • Obfuscation (SAFE): No Base64, zero-width characters, or other encoding techniques were found in the provided files.
  • Unverifiable Dependencies & Remote Code Execution (SAFE): The skill consists only of markdown files; no package managers (npm, pip) or remote script downloads (curl, wget) are present.
  • Privilege Escalation (SAFE): No commands requiring administrative privileges or system modification were identified.
  • Persistence Mechanisms (SAFE): No attempts to modify shell profiles, cron jobs, or registry keys were found.
  • Indirect Prompt Injection (LOW): The skill is designed to process user-provided contracts, which is an untrusted data source. While this presents a surface for indirect prompt injection, the skill lacks high-risk capabilities like file system writing or network requests, limiting the potential impact.
  • Ingestion points: User-provided contract text in 'Step 2 -- Analysis'.
  • Boundary markers: None explicitly defined in the workflow.
  • Capability inventory: Text generation and analysis only.
  • Sanitization: Not applicable as no code is executed.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 21, 2026, 11:45 AM
Security Audit — agent-trust-hub — assignment-general