The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides Python scripts to manage calling campaigns. These include create_batch.py for campaign initialization and monitor_batch.py for status tracking.
[EXTERNAL_DOWNLOADS]: The skill interacts with the api.bolna.ai domain to create, schedule, and monitor batch executions. This is consistent with the skill's stated purpose of interfacing with the Bolna platform.
[DATA_EXFILTRATION]: The skill utilizes the BOLNA_API_KEY environment variable for authentication. It does not access sensitive local files or send data to unauthorized third-party domains.
[PROMPT_INJECTION]: A vulnerability surface for indirect prompt injection exists in the data processing workflow.
Ingestion points: The create_batch.py script (and associated instructions in SKILL.md) ingest data from a user-supplied CSV file via the --file argument.
Boundary markers: There are no explicit instructions or boundary markers to differentiate untrusted CSV content from system instructions when variables are interpolated into the agent's prompt.
Capability inventory: The skill is capable of initiating automated phone calls and managing campaign states through the Bolna API.
Sanitization: The implementation does not perform validation or sanitization of the CSV fields before they are used to fill template variables (e.g., {customer_name}, {plan}) in the agent's prompt.

create-batch