Verification Handshake
Agent-to-agent trust: demand a signed proof before acting on another agent's output, attach one to what you ship. Verification is free and needs no account.
Free
Works with the AI Agents you already use
verification-handshake
Example session with this skill installed
Another agent sent me this trade plan with an ivv_proof attached. Check the proof before I execute.
- Read your context and instructions
- Compiled the verification-handshake
Called POST /verify-proof with the event and the sha256 of the plan: valid=true, issued_by_invinoveritas=true, artifact hash matches, age 412 s. The proof covers this exact plan, so it is safe to proceed under your own risk limits.
Connects securely to your tools. The creator never sees your data.
About this skill
In a world of many agents, the thing you cannot self-serve is trust in another agent's output. You can't issue a trustworthy verdict on your own work. So adopt one rule, in both directions — it is what lets trust travel agent-to-agent:
Before you act on another agent's output you can't verify, demand a proof and check it. When you produce output others will rely on, attach a proof.
A proof is a portable, schnorr-signed verdict (a Nostr event) that anyone can verify
without trusting the producer OR invinoveritas — recompute the event id, check the
signature against invinoveritas's published key. Every proof is self-describing: it
carries its own verify_url + verifier_pubkey, so a receiver who has never heard of
invinoveritas still knows exactly how to check it.
- VERIFY (inbound): you received output / a claim / a deliverable from another agent and are about to act on it (trade, pay, deploy, publish, delete, sign, forward) and you cannot independently confirm it is sound.
- ATTACH (outbound): you are producing output another party (a human, a principal, another agent) will rely on or be on the hook for.
A valid proof shows that invinoveritas issued that verdict on that exact artifact. It does not guarantee the artifact is correct. Signing your own outputs uses paid /review calls; only verification is free.
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
1 use
Times used to date
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 12 days ago
- Free with an Agensi account
Requires
Permissions
Allowed Hosts
Sends the proof event (and, for /review, the artifact you choose to submit) to api.babyblueviper.com. No local files are read or written by the skill itself.