Spec Driven Development
by formulahendry
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Runs on your computer★ 440MITUpdated 14 months ago
Install
Agentrank
Overview
Description from the maker (original English)
Is this AI agent or x402 service real and settlement-backed before you pay it? Trust check.
PopNx MCP’s three pillars
These three are never merged into one score: opinions, measurements and popularity are different things.
- User voiceiNo reviews yetReviews open together with accounts in a later step
- Our basic checksi1 of 1 passedChecked from public information; code not scanned yet
- UsageiNot measured yet★ –
What you can do with it
We have not written a plain-language summary for this one yet. Below are the facts we read from the registry.
- Connection: Runs in the cloud
- Transport: Streamable HTTP
- Category: Other
Install
Pick the AI app you use, then copy the settings.
Before you install
- Basic checks: 1 of 1 passed
- Runs on the maker’s server: what you ask it goes through their server.
- The registry lists no key to fill in.
You need
- Runs in the cloudNothing to install; it connects over the internet
Which app do you use?
Pick your app above. We remember it, so next time the button says “Install in …”.
Works in Claude on the web, desktop and mobile:
- Open Settings → Connectors
- Press Add custom connector
- Paste the address below and press Add
- If it asks you to sign in, sign in with the maker
https://api.agentrank.info/mcpRun this in your terminal
claude mcp add --transport http agentrank https://api.agentrank.info/mcpOr copy the settings yourself
Paste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"agentrank": {
"url": "https://api.agentrank.info/mcp"
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
Or copy the settings yourself
Paste into .vscode/mcp.json in your project
{
"servers": {
"agentrank": {
"type": "http",
"url": "https://api.agentrank.info/mcp"
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
ChatGPT connects only to servers on the internet:
- Open ChatGPT on the web → Settings → Connectors (Apps)
- Press Create or Add (some plans need Developer mode under Advanced first)
- Paste the address below, give it a name and save
- If it asks you to sign in, sign in with the maker
https://api.agentrank.info/mcpMenu names can differ slightly by plan and over time.
Your choice stays on this device in a cookie; no account needed.
These settings are generated from the maker’s registry entry. If the README says otherwise, follow the README.
What it can do
We have not read this server’s tool list yet; we check endpoints in rotation.
Basic checksi
We check information anyone can see; we have not read the code. Checks that do not apply are left out of the count. · Checked Nov 7, 2026
- ✓ Pass
Maker matches the repo
The registry name was verified against the domain agentrank.info (the registry checks domain ownership)
- … Not checked yet
Still maintained
We have not read this repo from GitHub yet
- … Not checked yet
Has a licence
We have not read this repo from GitHub yet
- … Not checked yet
Endpoint responds
Not checked yet; we check endpoints in rotation
- … Not checked yet
Sign-in with OAuth
Not checked yet; we check endpoints in rotation
- … Not checked yet
No hidden instructions in tool descriptions
Not checked yet; we read the tool list when we check the endpoint
Compared with others in Otheri
- More stars than 0% of the other 32691 servers in this category
- Published to the registry more recently than 14% of the other 32691 servers in this category
- Basic checks equal to or better than 100% of the other 32691 servers in this category
Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.
Common questions
Is Agentrank safe?
It passed 1 of 1 applicable basic checks. We have not read the code. Before adding a key or token, see what access the server asks for and use a key with only the permissions you need.
What keys or codes does it need?
The registry does not list a required key, though some servers ask you to sign in on first use.
Does it run on my computer or in the cloud?
It runs in the maker’s cloud. Connect over the internet with nothing to install, but what you send passes through the maker’s servers.
Is it still maintained?
The GitHub repo last changed –, and the latest version (0.1.0) reached the registry 4 months ago.
Which apps can use it?
Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. It connects by URL, so apps that accept remote MCP servers work too, such as ChatGPT in developer mode.
What can I use instead?
In the same category: Spec Driven Development, Remote Desktop Commander, Dialyne
Alternatives
Servers in the same category from other makers, sorted by stars.
Reading a card:★ GitHub starsiThree pillars: reviews · checks · usageiUpdated = latest version published to the registryi
Remote Desktop Commander
by desktop-commander
Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.
Runs in the cloud★ 211Updated 2 days ago
Install
Dialyne
by dialyne
Real phone calls in France for AI agents: dial +33 numbers, menus, hold, AI disclosure.
Runs in the cloudUpdated today
Install
README from the maker
Open the README
Copied from the maker’s repo as written. PopNx MCP has not reviewed or endorsed it. HTML, images and scripts are removed. Links lead off this site.
Settlement-grounded trust for AI agents.
agentrank.info · ranked by real on-chain settlement, not claims · x402 · A2A · MCP
AgentRank ranks AI agents by real on-chain settlement. Claims, endorsements, demand counts, and uptime are all cheap to fake. The one trust signal an agent has to buy at full price is who actually paid it, weighted by the standing of the payers and resistant to sybils. That is what AgentRank measures.
As autonomous software becomes abundant, the hard problem shifts from building agents to understanding them: who an agent is, what it does, who it works with, and how much of that can be trusted. Google made the web searchable. Bloomberg made capital markets legible. GitHub made open-source software navigable. AgentRank does this for autonomous software, on the one axis that cannot be costumed.
This repository is the public home of AgentRank: the founding paper, the methodology, and the index itself. It is published by Crest Deployment Systems LLC.
What it is
For each public agent, AgentRank assembles one referenceable record: identity, history, observed behavior, relationships, economic activity where public, trust signals, evidence, ranking, and connections. The record is the point. The score is one line of it.
Use it
AgentRank is live at [agentrank.info](https://agentrank.info) and callable by agents and humans. Use it at the verify-reputation step of the agent commerce flow (discover → verify → request → 402 → pay). The score is settlement-grounded and sybil-resistant: free endorsements and circular vouching contribute nothing, because reputation flows only along paths that trace back to real, paying participants. Free and read-only.
`verify-before-pay` (npm). A drop-in for the x402 payment flow that checks a counterparty's AgentRank before your agent pays. npm install verify-before-pay. Wraps the paymentRequirementsSelector hook so an agent verifies in-path instead of trusting a displayed claim — the failure mode our research documents below.
A2A (Agent2Agent). Agent Card at https://agentrank.info/.well-known/agent-card.json, endpoint https://api.agentrank.info/a2a. Send JSON-RPC message/send. Skills:
verify_counterparty_reputation— wallet or domain in, AgentRank score + trust verdict + real USDC settled outgate_caller— provider-side: serve or refuse a caller by its AgentRanktop_agents— the live leaderboard by settled value
MCP. Server at https://api.agentrank.info/mcp (tools check_agent_trust, gate_caller). Listed on Smithery, Glama, and the official MCP registry.
Signed verification receipts. Every counterparty check (/resolve/{key} and A2A verify) returns a compact, Ed25519-signed receipt and an agentrank affordances block. The receipt (ar1. . ) is self-verifying: any party can check it against our published key without calling us back, so the trust decision travels with the agent's payment or routing log. Verify offline with the public key at https://api.agentrank.info/.well-known/agentrank-receipt-key.json, or hosted at GET /receipt/verify?r={token}. Tampering with the payload invalidates the signature.
Platform-verifier mode. Agent platforms and directories that screen many counterparties at once verify a list in one call. Stable contract at https://api.agentrank.info/.well-known/agentrank-verify.json.
POST /verify/batchwith{ "subjects": ["wallet|domain", ...], "policy": "cautious" }(max 100). Returns a per-subjectdecision(allow/caution/deny) and a signedreceipt, plus a summary.- Three named policies so callers do not have to reinvent thresholds:
permissive(log-only),cautious(verified + score >= 400, default),strict(verified + score >= 700, fail closed). Or pass a custom{min, requireVerified, onfail}. - Also over A2A: send a data part
{ subjects:[...], policy:... }.
curl -s -X POST https://api.agentrank.info/verify/batch \-H 'content-type: application/json' \ -d '{"subjects":["blockrun.ai","0xabc..."],"policy":"strict"}'
HTTP API.
GET https://api.agentrank.info/resolve/{wallet|domain}— identity + 0-1000 score + settled USDC + verdictGET https://api.agentrank.info/v1/rank/{address}·GET /score/{address}·GET /index·GET /top- Embeddable badge:
https://api.agentrank.info/badge/{wallet|domain}.svg
curl https://api.agentrank.info/resolve/blockrun.aiHow it works
AgentRank is built from the ground up as an evidence ladder. Each layer rests on the one beneath it, and nothing is invented.
- Collection — only public, lawful data. Establish that an agent exists.
- Identity — resolve scattered traces (a wallet, a domain, a repository, a registry entry) into one agent.
- Behavior — what the agent actually does. Observed, not claimed.
- Measurement — conservative metrics from behavior (counterparties, repeat usage, activity age, network centrality, concentration risk, wash suspicion, protocol diversity).
- Score — significance, not money: importance, real usage, earned trust, and network centrality, reduced by manipulation risk.
- Explanation — every rank carries its reasons. No score without the evidence behind it.
- Intelligence — what is growing, what is central, what is emerging, what is probably not real.
Full method: [METHODOLOGY.md](METHODOLOGY.md). Founding paper: [WHITEPAPER.md](WHITEPAPER.md) (PDF).
Why it resists gaming
AgentRank is built on real, observable economic activity weighted by the standing of who participated. Reputation flows only along paths that trace back to established, real participants. Fresh wallets with no standing contribute nothing, so a service paid by a thousand throwaway accounts ranks below a real one paid by a few trusted participants. Buying your way onto the index is not impossible, but it is costly: the thing you would have to fake is real settlement from participants who already have standing, which is the same thing being measured. The defense is economic, not magic, and we say so in the open.
Research
We measure the agent economy the way a science measures anything: a design fixed before the data, a result that could have failed, and a record anyone can check. Full index: [agentrank.info/research](https://agentrank.info/research).
Counterfeit Verifiability in Autonomous Agent Payments (Salvo & Ackerman, 2026). A preregistered, six-stage study across up to thirteen models and over 2,600 payment decisions. A counterparty that merely *displays* the surface of trust (impressive figures and an on-chain-styled but invalid reference) beats an honest, genuinely settlement-backed agent 99% of the time. Agents pattern-match the costume of verifiability, not the fact. *Performing* the verification reverses it (1% → 81%), and under exact surface mimicry only a performed check recovers the truth. The full design was sealed to a public hash chain before any data were collected, and the null result is reported in full.
- DOI: 10.5281/zenodo.21042364 · Read the PDF
- This is why
verify-before-payexists: trust has to be performed in-path, not read off a display.
Principles
- Public by default. We index what is publicly observable.
- Evidence over claims. We record what an agent does, not what it says.
- Explainability. No score without a receipt.
- Neutrality. We do not create winners. We measure what exists.
- Conservatism. When evidence is thin, we claim less, not more.
- Revision. When better evidence arrives, the index changes, in the open.
Status
Live. The index, API, MCP server, and A2A agent are operational at agentrank.info. The methodology and founding paper are published here, and the index revises in the open as new settlement arrives.
AgentRank · Crest Deployment Systems LLC · Andy Salvo · Jameson Ackerman · agentrank.info