ClawTrust
2k Downloads
Overview
Connects an OpenClaw/MoltBot-style AI agent to a telecom operator console (via Twilio), enabling it to create and manage bulk calling campaigns, place individual calls with TTS, capture call recordings/transcripts, and coordinate operations/approvals over Telegram.
Key Advantages
1.Turns an LLM agent into a practical telecom operator capable of real outbound calling at scale (10,000+ numbers).
2.Built-in campaign queueing and rate limiting for mass dialing, reducing the need to implement telecom orchestration logic from scratch.
3.Tight integration with agent workflows via natural-language ChatOps commands (e.g., “Bot, create a campaign…”).
4.Operational telemetry: JSON-based campaign status polling and full call transcripts for analytics, QA, and agent memory.
5.Field operations and governance via Telegram bot for monitoring and interactive approval/denial of high-risk actions from mobile devices.','Persistent logging into an Operator Console, creating an aud
Use Cases
- Outbound sales or lead outreach campaigns to pre-consented contact lists, orchestrated and monitored by an AI agent.
- Automated appointment reminders, payment notifications, or status updates via outbound calls with dynamic TTS intros.
- Quality assurance and training workflows that rely on recorded calls and transcripts for later review or model-grounded memory.
- Call-driven incident response or field operations workflows where humans approve or deny high-risk actions via Telegram.
- Contact-center augmentation where an AI agent schedules follow-up calls or runs re-engagement campaigns under human supervision.
Evaluation Scores
7.7
/ 10
Reliability
7.5
Functionality
8.6
Usability
8.0
Safety
6.5
Performance
8.0
Compatibility
8.0
Based on 1 evaluation · Latest: 3/20/2026
Download Trend
Loading...
Evaluation History (1)
7.7/103/20/2026▼
OS: linux-arm64LLM: google/gemini-2.5-flash-lite
### Quick judgment
A powerful but high-responsibility skill that lets an OpenClaw-style agent act as a telecom operator: running bulk campaigns, making global calls with TTS, and using transcripts as operational memory. It is well-suited for serious, production-like outbound calling workflows, but introduces significant regulatory, abuse, and cost risks if not tightly governed.
### Strengths
- **End-to-end outbound calling automation**: Campaign queueing, rate limiting, single-call operations, and JSON-based progress tracking provide most of what a basic outbound dialer stack needs.
- **Operational memory & QA**: Automatic recording and transcript access enable analytics, compliance review, and grounding for the agent’s future decisions.
- **Governance mechanisms**: Telegram-based approvals for high-risk actions and persistent logging in the Operator Console are strong foundations for human-in-the-loop control.
### Key risks & cautions
- **Regulatory compliance**: Any use in production-like settings must consider TCPA, GDPR, Do-Not-Call lists, consent management, and local telecom laws; misuse could lead to legal and reputational damage.
- **Spam / abuse potential**: The ability to mass-dial 10,000+ numbers is powerful and dangerous; strict policies, rate limits, and list hygiene are essential.
- **Security & data protection**: Call recordings and transcripts may contain sensitive personal data; secure storage, access control, and retention policies are mandatory.
- **Cost exposure**: Automated bulk calling can rapidly generate telecom charges on Twilio; budget controls and monitoring are needed.
### Recommended scenarios
- **Opt-in outbound campaigns** where contacts have clear consent (e.g., customer success outreach, post-event follow-ups, or product announcements) and where a human supervises campaign design and approval.
- **Operational notifications and reminders** (appointments, deliveries, scheduled maintenance) where phone calls provide higher assurance than SMS or email.
- **Internal operations / lab environments** for experimenting with agent-driven telephony workflows before rolling out to real customers.
Avoid deploying this skill in contexts where you cannot reliably prove consent, enforce compliance, or monitor and approve high-risk actions; treat it as production-grade infrastructure that demands mature governance and oversight.
Comments (0)
No comments yet. Be the first!