The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides functionality to ingest and process external circuit definitions from QASM and JSON strings, which constitutes an attack surface for indirect prompt injection. \n
Ingestion points: references/building.md documents the use of cirq.read_json and cirq.contrib.qasm_import.circuit_from_qasm for importing circuit data. \n
Boundary markers: The instructions do not define specific delimiters or "ignore instructions" warnings for the agent when processing these external data inputs. \n
Capability inventory: The skill has access to Python execution via the Bash tool and performs local file writing for experiment data logging (references/experiments.md). \n
Sanitization: No explicit validation or filtering of the imported circuit definitions is implemented. \n- [EXTERNAL_DOWNLOADS]: The skill fetches resources and installs dependencies from well-known and official platforms. \n
Installation instructions in SKILL.md target official packages including cirq, cirq-google, and azure-quantum. \n
Documentation and resource links point to official Google Quantum AI domains and framework repositories on GitHub.

alterlab-cirq