generating-weaverse-project-json

Pass

Audited by Gen Agent Trust Hub on Oct 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted external data, including design specs, verbal descriptions, and existing exports, to generate structured JSON. This represents an attack surface where malicious instructions could be embedded in the input data.\n
  • Ingestion points: SKILL.md lists inputs such as design specifications, verbal descriptions of storefronts, and existing project exports.\n
  • Boundary markers: The instructions do not specify the use of delimiters or specific warnings to ignore instructions within the input data.\n
  • Capability inventory: The skill's primary capabilities are reading local project files (app/sections/, app/components/) and executing the local scripts/validate.py script.\n
  • Sanitization: The skill does not explicitly describe any sanitization or validation of the input content before processing.\n- [DYNAMIC_EXECUTION]: The agent is instructed to execute a local Python script for data validation.\n
  • Evidence: SKILL.md directs the agent to run python scripts/validate.py <output-file>. An audit of the script shows it is a benign utility that performs structural integrity checks on the JSON output and has no network or privilege-escalation capabilities.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 5, 2026, 07:18 PM