- Home
- Skills
- Agents & Orchestration
- Citizen Services Voice Agent
Citizen Services Voice Agent
Scaffolds a multilingual phone agent for public services, with caller authentication and answers grounded in official documents.
Free
Works with the AI Agents you already use
About this skill
Builds a multilingual phone voice agent for public-sector citizen services. The coding agent first asks for the agency's services, call languages and identity factors, then scaffolds a project with these parts.
- ElevenLabs Conversational AI agent with a workflow that keeps caller data behind a webhook-based identity check
- LangGraph backend served as an OpenAI-compatible custom LLM over SSE, with one node per service intent and status-dependent lookups chained in code
- Retrieval over official documents with language and category filters and a relevance threshold
- Signed session tokens, keyed hashing of identifiers, redacted logging, and a circuit breaker with a health endpoint
- Contract tests for the SSE endpoint and a validator for the knowledge base manifest
The skill contains eleven reference documents, among them the custom LLM contract, caller authentication, knowledge base design, voice output rules, ElevenLabs configuration and platform behavior observed in practice, plus tested code templates.
The generated project needs the user's own ElevenLabs, OpenAI and Pinecone accounts, and a phone number provider such as Twilio for phone calls.
Based on the author's open demo project government-citizen-services-voice-agent, in English and Turkish. The source is at https://github.com/umutdinceryananer/government-citizen-services-voice-agent and a walkthrough video is at https://youtu.be/kSFCPI2FEMs
Based on a demo project that has not run in production. Identity factors, data retention and deployment must follow the deploying agency's policy. ElevenLabs field names and behavior change between releases, so the platform notes in the skill need re-checking. The voice platform needs a public HTTPS host for the backend.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 4 days ago
- Free with an Agensi account
Requires
Permissions
Allowed Hosts
The skill writes a project into the working directory and runs its tests in a shell. The generated project calls the ElevenLabs, OpenAI and Pinecone APIs with the buyer's own keys from the project's settings. The files in the skill make no network calls and read no environment variables.