The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to run a local script scripts/score.py using python3 to evaluate a generated draft.json file. While functional, this involves executing code on the local system.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input in the form of 'buyer questions', which could be used as a vector for indirect prompt injection. 1. Ingestion points: The buyer question written in step 1 of the checklist. 2. Boundary markers: No delimiters or warnings are used to differentiate user content from instructions. 3. Capability inventory: The skill has the capability to execute a local Python script. 4. Sanitization: No sanitization or filtering of the user-provided text is specified.

buyer-question