The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides installation instructions for various Python packages (pennylane, pennylane-qiskit, pennylane-cirq, etc.) from official and established quantum computing service providers.
[DYNAMIC_EXECUTION]: Uses the Catalyst framework for Just-In-Time (JIT) compilation through the @qjit decorator to optimize the execution speed of quantum circuits, which is a standard performance feature in the scientific domain.
[INDIRECT_PROMPT_INJECTION]: The skill contains instructions for the agent to fetch citation metadata from external URLs (arXiv) to ensure accurate referencing. This creates a data ingestion surface for external content.
Ingestion points: Citation metadata fetched from https://arxiv.org/abs/2609.00065 via network tools.
Boundary markers: No specific delimiters or warnings for the fetched content are defined in the instructions.
Capability inventory: The skill possesses Read, Bash, and Python capabilities that could interact with the processed data.
Sanitization: No explicit sanitization or validation of the fetched metadata is performed before its use in the citation.
[COMMAND_EXECUTION]: Includes uv pip install commands for environment setup and provides Python-based examples for running quantum simulations and optimization loops.

pennylane