The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python script rebalance.py to calculate portfolio drift and generate trade proposals based on user-defined parameters.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from several local JSON files which could potentially contain malicious instructions if the tools generating those files are compromised.
Ingestion points: The script reads data from trader/portfolio.json, trader/prices.json, trader/targets.json, and trader/config.json via the read_json and parse_targets functions.
Boundary markers: Output is formatted as a drift table for the agent to display, but the skill does not use specific delimiters to isolate external data from its own instructions.
Capability inventory: The skill is capable of reading and writing local files within the trader/ directory. It does not perform network operations or execute arbitrary subprocesses.
Sanitization: Input data is parsed as JSON and numerical values are cast to floating-point numbers or integers, which mitigates many common injection vectors by stripping non-numerical content from calculations.

trader-rebalance