The Agent Skills Directory

[SAFE]: The skill provides instructional content for porting quantum computing code. All analyzed patterns, including command-line examples and code snippets, are standard for the intended domain of quantum circuit development.
[SAFE]: The skill includes references to official NVIDIA GitHub pages and documentation (nvidia.github.io, github.com/NVIDIA/*), which are recognized as trusted vendor resources.
[SAFE]: The python -c execution example in SKILL.md is a benign version check for the cudaq library.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided Qiskit source code (ingestion point: SKILL.md workflow). While this presents an attack surface where a user could provide a malicious circuit, the skill does not grant the agent dangerous capabilities like arbitrary file writing or non-whitelisted network access. The agent is guided to produce code output, which is then subject to standard user review.

cudaq-importing