The Agent Skills Directory

[DYNAMIC_EXECUTION]: The commands get_mock_tick and get_plays_near_timestamp accept a mock_file_path parameter, allowing the skill to load and process data from local file paths provided at runtime. This functionality enables dynamic loading from computed paths, which could be misused if an agent is directed to access a sensitive system file.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sports and prediction market providers (ESPN, Kalshi, Polymarket), creating a potential vector for indirect prompt injection.
Ingestion points: Data enters the agent context through API calls in get_todays_markets, search_entity, and match_markets.
Boundary markers: No specific delimiters or instructions are provided to the agent to treat external content as untrusted or to ignore embedded instructions.
Capability inventory: The skill is capable of price comparison, odds normalization, and arbitrage evaluation, but it lacks tools for writing files or executing shell commands based on the external input.
Sanitization: There is no evidence of validation or sanitization of the external API content before it is utilized by the agent for decision-making.

markets