The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill interacts with 'api.bolna.ai', which is the official API endpoint for the skill's author ('bolna-ai'). These interactions are necessary for the skill's functionality (creating and testing dispositions).
[COMMAND_EXECUTION]: The provided Python scripts ('bulk_create_dispositions.py', 'create_disposition.py', 'test_dispositions.py') perform standard HTTP requests using the built-in 'urllib' library to communicate with the Bolna API. There is no evidence of arbitrary or dangerous shell command execution.
[CREDENTIALS_UNSAFE]: The skill correctly instructs the user to provide the 'BOLNA_API_KEY' via an environment variable. It does not contain any hardcoded secrets. Standard placeholder strings like 'REPLACE_WITH_YOUR_AGENT_UUID' are used in asset files.
[INDIRECT_PROMPT_INJECTION]: While the skill processes transcripts and defines LLM prompts (dispositions), it serves as a management tool for these configurations. The extraction process itself is performed by the Bolna platform, and the skill includes documentation on using specific validation types (regex, email, etc.) to ensure data integrity.

create-disposition