Skills
Navigate
skills/lawve-ai/awesome-legal-skills/outside-counsel-billing-performance-reviewer/Gen Agent Trust Hub

outside-counsel-billing-performance-reviewer

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted external data in the form of billing narratives and invoice details provided by third-party law firms. This creates a surface for indirect prompt injection where malicious instructions embedded in invoice narratives could attempt to override the agent's audit logic.
  • Ingestion points: scripts/normalize_billing_data.py and scripts/xlsx_utils.py read data from various file formats including CSV, XLSX, JSON, and LEDES.
  • Boundary markers: Absent. The skill instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings for the processing of third-party narrative text.
  • Capability inventory: The skill can execute Python scripts to write to the file system, specifically generating issue logs via scripts/export_issue_log.py and executive reports via scripts/build_exec_pack.py.
  • Sanitization: Present. The skill implements sanitization using html.escape in scripts/build_exec_pack.py and _xml_escape in scripts/xlsx_utils.py to prevent cross-site scripting (XSS) or XML injection in generated artifacts.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 07:33 PM
Security Audit — agent-trust-hub — outside-counsel-billing-performance-reviewer