The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's primary workflow requires the agent to run a local Python script (scripts/position_sizer.py) to perform financial calculations. This execution is the intended purpose of the tool and is performed locally without external dependencies.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied trade parameters, creating a potential surface for indirect injection.
Ingestion points: User data (account size, entry/stop prices) is ingested via command-line arguments in the scripts/position_sizer.py script.
Boundary markers: The instructions do not define specific delimiters for user data, relying on the agent's natural language processing to extract numerical values.
Capability inventory: The script can create directories and write JSON and Markdown reports to the local file system using os.makedirs and open().
Sanitization: The tool includes a validate_parameters function that enforces strict numerical constraints, such as ensuring positive values and logical price offsets, which mitigates the risk of malformed or malicious data affecting the calculations.
[DYNAMIC_EXECUTION]: The test suite in scripts/tests/test_position_sizer.py uses subprocess.run to verify the command-line interface of the tool. This usage is restricted to testing the skill's own components and does not introduce external execution risks to the agent's environment.

position-sizer