contract-delivery-kickoff

Pass

Audited by Gen Agent Trust Hub on Aug 5, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The scripts/contract_kickoff.py script utilizes PowerShell to interface with Excel through COM objects (via the _VERIFY_BODY script block). This functionality is intended for verifying workbook calculations. The inputs to these commands are derived from internal logic and numeric data, minimizing the risk of arbitrary command injection.
  • [PROMPT_INJECTION]: The skill processes external Word documents (.docx), which introduces an indirect prompt injection attack surface.
  • Ingestion points: Untrusted data is ingested from contract tables and prose in scripts/contract_kickoff.py via the load_document function.
  • Boundary markers: While explicit boundary markers are not present, the skill utilizes specific keyword-based extraction (e.g., searching for budget headers) to isolate data.
  • Capability inventory: The skill has capabilities for file system access (read/write) and shell command execution via PowerShell.
  • Sanitization: Extracted strings are normalized and validated using regular expressions (e.g., parse_number) before being processed, which limits the potential for executing embedded instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 5, 2026, 01:42 PM
Security Audit — agent-trust-hub — contract-delivery-kickoff