The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill makes network requests to api.bolna.ai (official vendor endpoint) and defaults to api.openai.com for LLM services. These are legitimate operations for the skill's purpose.
[COMMAND_EXECUTION]: The skill includes a Python script (scripts/create_minimal_agent.py) that uses standard libraries like argparse and urllib to communicate with the Bolna API. No dangerous command execution or shell injection patterns were found.
[CREDENTIALS_UNSAFE]: The skill correctly instructs users to set the BOLNA_API_KEY environment variable and does not contain hardcoded secrets. It follows best practices for secret management by accessing the key via os.environ.get("BOLNA_API_KEY").

create-agent