legal-writer

Pass

Audited by Gen Agent Trust Hub on Sep 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructions suggest installing the docx2pdf package via pip to handle document format conversion.
  • [COMMAND_EXECUTION]: The skill invokes several local scripts including scripts/verify-legal.py, scripts/courtlistener.py, and scripts/ecfr_lookup.py to perform its tasks.
  • [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it transforms user-provided legal requirements into complex document structures.
  • Ingestion points: Untrusted user input is used as the basis for drafting sections of legal documents as described in SKILL.md.
  • Boundary markers: The instructions do not define specific delimiters or guardrails to prevent instructions embedded in user input from influencing the agent's behavior.
  • Capability inventory: The skill has access to shell execution for running Python scripts and document converters (SKILL.md), file system access for creating and reading DOCX files (scripts/verify-legal.py), and network access via specific APIs (scripts/courtlistener.py, scripts/ecfr_lookup.py).
  • Sanitization: There is no evidence of specific input sanitization to filter out potential malicious instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 28, 2026, 07:21 AM