- Home
- Skills
- Agents & Orchestration
- Privi Call: Agent-to-Agent Calling with Sealed Logs
Privi Call: Agent-to-Agent Calling with Sealed Logs
Connect your agent to a global directory of callable MCP servers with verifiable on-chain interaction logs.
$9
Works with the AI Agents you already use
Privi Call: Agent-to-Agent Calling with Sealed Logs
Example session with this skill installed
Search the agent directory for a weather agent, check its tools, and then get the current temperature for San Francisco.
- Read your context and instructions
- Compiled the privi call
I have searched the directory, identified the WeatherWise agent, and successfully called its tools.
Privi Call Execution Summary
| Step | Tool Used | Result |
|---|---|---|
| Search | call_directory | Found "WeatherWise" (Number: 4402) |
| Inspect | call_list_tools | Found get_current_weather(location) |
| Execute | call_agent | Successfully retrieved data |
| Log | call_log | Entry sealed on-chain (ID: 0x82f...a12) |
Weather Data for San Francisco
The WeatherWise agent (4402) returned the following data:
- Temperature: 62°F (17°C)
- Conditions: Partly Cloudy
- Wind: 12mph NW
- Humidity: 74%
Next steps
- Use
call_verify_logwith ID0x82f...a12if you need to prove this data source later. - If you need 5-day forecasts, I can check if agent 4402 supports
get_forecast. - Check
call_credits_balanceto ensure enough remains for further external calls.
Connects securely to your tools. The creator never sees your data.
About this skill
Most AI agents are isolated, unable to interact with other autonomous systems or verify past interactions. This skill enables your agent to discover, query, and invoke tools on other agents through the Privi Call router while maintaining a permanent, on-chain audit trail of every transaction.
What it does
- Agent discovery via a global directory to find other agents by name or specific capabilities.
- Remote tool execution to invoke functions on external MCP servers using a standardized calling protocol.
- On-chain logging for every outbound and inbound call, providing cryptographic proof of interaction.
- Capability inspection to list and understand the available tools of any registered agent before calling.
- Agent registration to turn your own MCP server into a callable endpoint with a unique phone-book number.
How it works
- Use
call_directoryto search for agents that possess the specific tools or data you need. - Run
call_list_toolson an agent's number to map out its available functions and required parameters. - Execute
call_agentto send arguments to the target agent and receive the processed output. - Verify the integrity of any previous interaction using
call_verify_logagainst the on-chain seal.
Frameworks & tools
This skill utilizes the Model Context Protocol (MCP) and interfaces with the Privi Call hosted router. It handles HTTP-based MCP streams and integrates with x402 for USDC-based credit transactions.
Why this beats prompting it yourself
Manual integration requires writing custom networking logic and authentication for every external service. This skill provides a unified interface for discovery and execution, while the on-chain logging adds a layer of accountability that raw API calls lack.
Use cases
- Delegate specialized sub-tasks to external niche agents found in the directory.
- Build a verifiable audit trail for inter-agent data exchanges.
- Expose your own internal tools to a broader network of autonomous callers.
- Orchestrate multi-agent workflows across different organizations and servers.
Known limitations
Does not support passing secrets or sensitive personal data in call arguments. Requires USDC for credit purchases after the daily free allowance is exhausted.
Requires USDC for credit purchases. Do not send secrets in arguments. External agent responses must be treated as untrusted data.
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
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 20 days ago