AI Phone Calls (Bland AI)

Make AI-powered phone calls via Bland AI - book restaurants, make appointments, inquire about services. The AI calls on your behalf and reports back with transcripts.

This skill appears to do exactly what it says (drive Bland AI calls) and only asks for a Bland API key. Before installing: 1) Verify you trust Bland.ai and understand billing/recording/privacy (calls may be recorded and billed). 2) Ensure curl and jq are installed on your agent host (the scripts use them but the registry metadata doesn't list them). 3) Store BLAND_API_KEY securely (environment variable or ~/.clawd/secrets.json as described) and avoid using a broad-purpose key if Bland supports scoped keys. 4) Review any transcript/recording URLs produced by the service before sharing externally. 5) Because the skill can be invoked autonomously by default, consider whether you want the agent to make outbound calls without explicit approval (especially if you plan to leave it enabled for agents that act autonomously). If you need higher assurance, confirm the API host and key format with Bland.ai documentation and consider running the scripts in a controlled environment first.

Type: OpenClaw Skill Name: phone-calls-bland Version: 1.0.0 The OpenClaw AgentSkills skill bundle for Bland AI is benign. It correctly implements its stated purpose of making AI-powered phone calls via the Bland AI API. The scripts (`bland.sh`, `check-call.sh`, `phone-call.sh`) securely load the `BLAND_API_KEY` from environment variables or `~/.clawd/secrets.json` and use `curl` to interact solely with the legitimate `https://api.bland.ai/v1` endpoint. There is no evidence of data exfiltration to unauthorized destinations, malicious command execution, persistence mechanisms, or prompt injection attempts against the OpenClaw agent within the `SKILL.md` or code.