The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized data access were detected. The skill's behavior matches its stated purpose of providing a local simulated trading ledger.
[COMMAND_EXECUTION]: The skill executes its local entry script run.py using python3. This is the intended implementation for its trade simulation logic and is constrained to the skill's own package.
[DATA_EXPOSURE]: The skill reads and writes to a local JSON file to persist the paper-trading ledger. This is restricted to the local runtime area and does not involve sensitive system files or credentials.
[INDIRECT_PROMPT_INJECTION]: The skill has an ingestion surface for untrusted data through command-line arguments and market data files.
Ingestion points: User-provided arguments (e.g., --note, symbol names) and external market data files (e.g., --csv).
Boundary markers: None explicitly defined in the prompt instructions; reliance is placed on the underlying parser.
Capability inventory: The skill uses the Bash tool to run its local Python simulation script.
Sanitization: The skill utilizes a structured argument parser from its support library (core.skill_support) which helps mitigate direct command injection via arguments.

paper-trading