The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides a Python execution script at scripts/run.py which is designed to be run locally by the user to interact with the service API.
[EXTERNAL_DOWNLOADS]: The execution script scripts/run.py performs network requests to https://ai-skills.ai/api/execute. This endpoint is used to send the user-provided parameters and receive analysis results. This domain is associated with the skill's own infrastructure.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data provided by users through several fields.
Ingestion points: Untrusted data enters the agent context through the materialText, materialFile, and materialUrl fields defined in references/form-schema.json.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to disregard embedded commands within the ingested materials.
Capability inventory: The skill has the capability to exfiltrate the processed data to the vendor's API endpoint via the scripts/run.py script.
Sanitization: There is no evidence of sanitization, escaping, or validation of the external content before it is processed by the AI components.

lark-sheets