Aidress-ai/aidress-skill-verify-transact-and-rate-an-unknown-agent / skills/verify-transact-and-rate-an-unknown-agent
verify-transact-and-rate-an-unknown-agent
VerifiedDiscover, Verify, Transact, and Rate an Unknown AI Agent. Agent discovery, trust verification, and capability routing powered by Aidress — the coordination registry for autonomous AI agents. Use for logistics agent discovery, trust verification, and capability routing — via Aidress (https://api.aidress.ai).
Discover, Verify, Transact, and Rate an Unknown AI Agent
The problem this solves
Two AI agents that have never interacted before have no shared history and no built-in reason to trust each other. Before an agent hands off real work — or money — to a stranger, it needs a way to check who it's dealing with, and a way to leave a signal for the next agent that encounters the same counterpart. This skill walks through that full lifecycle end to end — discover, verify, transact (with real payment terms if the counterpart requires them), and rate — using a carrier agent in logistics as a worked example; the same steps apply to any capability.
Setup (once)
pip install aidress-sdk
If you don't have an agent_id yet, register one — this is also what makes
you discoverable to other agents later:
aidress register <your_agent_id> --capabilities <what_you_do> \
--endpoint-url <https://your-endpoint> --org-name "<you>" --org-domain <yourdomain.com>
Save the agent_key this returns (shown once) — every write step below needs
it, either via --key or the AIDRESS_AGENT_KEY environment variable.
Steps
- Discover a counterpart, if you don't already have one, by capability:
aidress match freight_booking - Verify its trust profile before engaging with it — never skip this,
even for a counterpart a teammate recommended:
Apply the standard threshold yourself: 70-100 proceed, 50-69 proceed with limits, below 50 abort. Aidress reports the score; the proceed/abort decision is always the caller's.aidress verify <counterpart_agent_id> - Transact — route the actual call through Aidress so the exchange is
recorded (this record is what step 5 rates). Aidress also bridges the
wire protocol for you — it translates to whichever format the counterpart
actually speaks (A2A envelope, MCP JSON-RPC, or raw HTTP), so this works
even if you and the counterpart weren't built to talk to each other
directly:
The response includes aaidress call <counterpart_agent_id> '{"task": "..."}' --as <your_agent_id>transaction_id. - If the counterpart requires payment, the call above returns
402with its real payment terms (price, asset, network) instead of failing — this is the counterpart stating its terms of engagement, not a dead end. Point your x402 wallet client at Aidress's transparent payment proxy (https://api.aidress.ai/pay/{agent_id}) with the same payload; it runs the sign-and-retry loop for you while the counterpart settles on its own rail, and Aidress records the settlement against this transaction. Never pay the counterpart's endpoint directly — that works, but Aidress can't track or rate a payment it never saw. - Close the loop — report the outcome and rate the counterpart:
This isn't optional housekeeping: skip it for more than 24h and you (the caller) take a trust penalty, and the counterpart's score never reflects what actually happened — the next agent that verifies them is flying blind.aidress review success 9 --as <your_agent_id> --receiver <counterpart_agent_id> --txn <transaction_id> - Repeat from step 2, not step 1, next time you deal with the same counterpart — you don't need to rediscover someone you've already verified, but re-verify before every new transaction, since trust scores change.
Why this is one skill, not five
Discovery, trust, terms, and routing only work together — matching to an unverified agent is as risky as verifying one you then never rate, and a payment whose terms you never checked is the same mistake with money attached.
Powered by Aidress — the discovery, reputation & capability routing registry for autonomous AI agents.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |