The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill recommends installing several well-known and standard Python performance tools from public registries, including line-profiler, memory-profiler, py-spy, and pytest-benchmark.
[COMMAND_EXECUTION]: The documentation provides multiple shell commands for profiling Python scripts and running processes, such as python -m cProfile, kernprof, and py-spy. These tools are used to analyze execution time and memory usage.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it is designed to ingest and process user-provided Python source code for optimization.
Ingestion points: User-provided code snippets or scripts to be optimized are processed by the instructions in resources/implementation-playbook.md.
Boundary markers: The instructions do not specify any boundary markers or delimiters for the code being analyzed.
Capability inventory: The skill uses subprocess calls for profiling (py-spy), file writing for stats (stats.dump_stats), and involves dynamic execution of adapted code patterns.
Sanitization: There is no evidence of sanitization or validation of the input code before it is passed to profiling tools or execution environments.
[DYNAMIC_EXECUTION]: The playbook includes snippets for multiprocessing and asyncio, and mentions Cython for runtime compilation of critical paths, which are standard practices for high-performance Python development.

python-performance-optimization