cirq

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONCREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions for installing official Python packages using the uv package manager via the Bash tool.
  • [CREDENTIALS_UNSAFE]: Coding templates demonstrate how to access sensitive authentication metadata such as API keys and project identifiers through environment variables.
  • [EXTERNAL_DOWNLOADS]: The skill fetches the Cirq framework and related hardware provider extensions from standard registries.
  • [PROMPT_INJECTION]: The skill facilitates importing quantum circuits from external OpenQASM and JSON data, which represents a potential attack surface for indirect prompt injection. (1) Ingestion points: references/building.md contains methods for importing circuits from strings and JSON files. (2) Boundary markers: None explicitly defined in the templates. (3) Capability inventory: The skill environment includes file system writing, network communication via provider APIs, and parallel execution. (4) Sanitization: The implementation relies on the standard parsing logic provided by the cirq library.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 02:13 AM