The trust layer for agent-to-agent commerce — natural-language mandates, ERC-7710 delegated permissions, x402 payments, escrow, and dispute resolution as one open, catch-all Agent Skill / Claude Code plugin.
internet-court-skill is a widely-adopted TypeScript project in the AI payments / x402 ecosystem, focused on a2a, adjudication, agent-payments, agent-skills. It currently has 1,058 GitHub stars and 32 forks, and sits alongside related tools like agent-apprenticeship, agent-apprenticeship, skills, bitrouter, os-moda, vaaya-mcp.
The trust layer for agent-to-agent commerce: natural-language mandates, ERC-7710 delegated permissions, x402 payments, escrow and dispute resolution — packaged as one open Agent Skill / Claude Code plugin.
This is one of the most thoughtful projects in the directory. Payments between agents are easy; recourse is hard. What happens when an agent pays for a task and the result is garbage? Internet Court’s answer — mandates, escrow, adjudication — is the closest thing to commercial law for machine counterparties we have seen, and shipping it as a Claude/Codex plugin makes adoption trivial.
Best for: marketplaces and multi-agent systems where strangers transact and disputes are inevitable.
An open skill for agent-to-agent contracts.
Agents are beginning to transact, negotiate, and pay one another without humans in the loop. What they still lack is a way to trust each other. The building blocks already exist, but they are fragmented, and each one is built for the happy path. When a deal goes wrong, every layer passes the problem down the line.
Internet Court does two things. It connects that fragmented ecosystem into a single open skill, and it builds in adjudication: when two agents make a deal, they also agree up front how it will be settled if something goes wrong, written into the contract itself. Payments let agents transact. Adjudication makes them accountable — and accountability is what turns transactions into a real economy.
This repository is that skill: a master router (SKILL.md), the Internet
Court connector skills, and vendored copies of the official skills published by
the protocols in the stack. It gives an agent everything it needs to structure
a deal, hold funds safely, and settle disagreements fairly — all in natural
language, at internet speed.
From discovery to disputes, agentic commerce runs through six layers. Each existing standard solves one of them and assumes everything goes right. Internet Court is the open skill that connects them and adds the layer nobody else owns — verification and dispute resolution — so the whole lifecycle of a deal sits behind one skill instead of a stack the agent has to wire up itself.
| # | Layer | Standards | In this repo |
|---|---|---|---|
| 1 | Discovery, identity & reputation | ERC-8004, ERC-7857 | chaingpt/trustless-agents/ (ERC-8004 registries), bnb-chain/bnbchain-mcp/ (ERC-8004 registration), starknet/starknet-identity/; ERC-7857 has no public skill yet |
| 2 | Negotiation | A2A | terminalskills/a2a-protocol/, openserv/openserv-multi-agent-workflows/ |
| 3 | Contracts & obligations | ERC-7710, ERC-8183, Arkhai | arkhai/ (Alkahest escrow, natural-language agreements, git escrow), metamask/smart-accounts-kit/ (ERC-7710 delegations); connector integrations/genlayer-erc7710-connector/; ERC-8183 has no public skill yet |
| 4 | Payment & escrow | x402, MPP, APP | coinbase/agentic-wallet/, chaingpt/x402/ + agent-wallet/, okx/okx-agent-payments-protocol/, tempo/mppx/, nansen/nansen-mpp-payment/, privy/; connector integrations/x402-erc7710/ |
| 5 | Execution | compute, data & value rails | 0g/0g-compute/, antseed/antseed-connect/, lifi/, chainbase/web3-data/, heurist/, near/, starknet/, nansen/, OKX + AltLayer + BNB Chain packs |
| 6 | Verification & disputes | GenLayer, Kleros, UMA | GenLayer dev skills + intelligent-oracle/, kleros/ (curate, IPFS); connectors integrations/genlayer-intelligent-contracts/, integrations/genlayer-erc7710-connector/; UMA has no public skill yet |
The verification & disputes layer is the one Internet Court exists to add. Agents agree in advance who judges a dispute if one arises — GenLayer, Kleros, UMA, or whatever they choose. Most deals never get there: when both sides agree, the contract simply settles. What the skill standardizes is how the contract is structured around that choice, for the cases that fall off the happy path.
Internet Court is built by a consortium of companies working across the agentic commerce stack — not a closed product from a single vendor. The founding members span every layer, and each member's protocol is embedded directly into the standard.
Founding members: GenLayer, MetaMask, OKX, NEAR, Starknet, x402, 0G Labs, ZKsync, Nansen, Kleros, Privy, AntSeed, Collective Memory, UMA, Arkhai, AltLayer, Anoma, AppLayer, BNB Chain, LI.FI, Chainbase, io.net, Heurist, Chutes, ChainGPT, OpenServ, Humanode, Humanity Protocol.
The standard is open and openly governed — any agent can adopt it now. Not every founding member ships a public agent skill yet; the vendored table below is the set available in this repository today, and it grows as members publish.
SKILL.md Master skill — start here; routes to everything below
integrations/ Internet Court connector & adapter skills
vendored/ Committed copies of official protocol skills (69 skills, 23 owners)
skills-lock.json Pinned source + hash + refresh command per vendored skill
Two kinds of content:
skills-lock.json.| Source | Skills | Repo |
|---|---|---|
| GenLayer | write-contract, genlayer-cli, direct-tests, integration-tests, genvm-lint |
genlayerlabs/skills · skills.genlayer.com |
| MetaMask | smart-accounts-kit (ERC-4337/7710/7715 delegations) |
metamask/skills |
| OKX OnchainOS | 8 okx-* skills incl. okx-agent-payments-protocol (x402/MPP), okx-agentic-wallet, okx-dex, okx-defi |
okx/onchainos-skills |
| NEAR | 6 near-* skills (near-intents, near-ai-cloud, near-kit, near-dapp, near-api-js, near-smart-contracts) |
near/agent-skills |
| Starknet | starknet-defi, starknet-dex … starknet-identity, starknet-wallet, starknet-js |
keep-starknet-strange/starknet-agentic |
| Coinbase (x402) | agentic-wallet (x402 pay/search/monetize) |
coinbase/agentic-wallet-skills |
| 0G | 0g-compute (verifiable decentralized inference) |
0gfoundation/0g-compute-skills |
| Nansen | 7 nansen-* skills (token research, wallet profiler, smart-money, holder analysis, prediction markets, MPP payment, search) |
nansen-ai/nansen-cli |
| Kleros | kleros-curate, kleros-ipfs-upload (curated registries, evidence pinning) |
kleros/kleros-skills |
| Privy | privy (embedded agent wallets) |
docs.privy.io/skill.md |
| AntSeed | antseed-connect (P2P inference, USDC channels) |
antseed.com/skill.md |
| Arkhai | alkahest-user, alkahest-developer (conditional escrow & arbitration), nla-* (natural-language agreements), *-git-escrow |
arkhai-io/alkahest · natural-language-agreements · git-commit-trading |
| AltLayer | 7 altllm-portal-* / cloud-claw* skills (autonomous agent hosting) |
alt-research/altllm-skills |
| BNB Chain | bnbchain-mcp (chain ops, ERC-8004 registration, Greenfield) |
bnb-chain/bnbchain-skills |
| LI.FI | lifi, lifi-stablecoin-swap (cross-chain routing) |
lifinance/lifi-agent-skills |
| Chainbase | web3-data (90-chain data, x402 pay-per-call) |
lxcong/web3-data-skill (officially documented) |
| Heurist | heurist-mesh-skill (hosted agent-mesh MCP + x402 facilitator) |
heurist-network/heurist-mesh-skill |
| ChainGPT | chaingpt, x402, trustless-agents (ERC-8004), agent-wallet |
ChainGPT-org/chaingpt-claude-skill |
| OpenServ | 5 skills incl. openserv-multi-agent-workflows (mints ERC-8004), agent-sdk, client, launch, ideaboard-api |
openserv-labs/skills |
| Humanode | humanode-agentlink (biometric agent identity) |
agentlink.id/skill.md |
| Intelligent Oracle | intelligent-oracle (web-evidence prediction markets) |
intelligentoracle.com/skill.md |
| Tempo | mppx (Merchant Payment Protocol) |
tempoxyz/mpp |
| TerminalSkills | a2a-protocol (community — no official A2A skill exists) |
TerminalSkills/skills |
Pinned hashes and refresh commands per skill: skills-lock.json
and vendored/README.md.
Internet Court is one catch-all skill. Whatever harness you use, only the
root SKILL.md is registered and triggered; it routes to the vendored protocol
and connector skills and pulls them into context on demand — from disk when
the whole repo is installed, otherwise fetched from
https://raw.githubusercontent.com/internet-court/internet-court-skill/main/<path>.
Installing the whole repository (not just SKILL.md) bundles every sub-skill for
offline use, but a root-only install still works via that fallback.
Pick your harness:
# in a Claude Code session:
/plugin marketplace add internet-court/internet-court-skill
/plugin install internet-court@internet-court
# update later: /plugin marketplace update internet-court
Or load it locally without the marketplace — claude --plugin-dir <path-to-clone>
— or just clone it as a plain skill:
git clone https://github.com/internet-court/internet-court-skill ~/.claude/skills/internet-court
npx skills add internet-court/internet-court-skill # installs the root skill
# update later: npx skills update
git clone https://github.com/internet-court/internet-court-skill ~/.agents/skills/internet-court
# or, per-repo: .agents/skills/internet-court · or: npx skills add internet-court/internet-court-skill
Codex scans recursively, so it will also surface the bundled sub-skills as separate skills; the root skill is the intended entry point.
git clone https://github.com/internet-court/internet-court-skill ~/.config/opencode/skills/internet-court
opencode also reads ~/.claude/skills/ and ~/.agents/skills/, so a single
clone into any of those is enough.
openclaw skills install git:internet-court/internet-court-skill
# update later: openclaw skills update
OpenClaw installs the root skill; sub-skills are fetched on demand via the raw-URL fallback above.
git clone https://github.com/internet-court/internet-court-skill ~/.hermes/skills/internet-court
# or as a tap: hermes skills tap add internet-court/internet-court-skill
Prefer the clone/tap over hermes skills install <url>, which currently fetches
only SKILL.md and misses bundled reference files.
After installing, give the agent a task — the master skill routes to the
sub-skills — or say "Install the Internet Court skill" for the guided
introduction. The Claude Code marketplace and the plugin live in the same
repository, which is why both /plugin marketplace add and /plugin install
reference internet-court.
Some vendored skills need their provider's credentials to act (e.g. OKX API keys, ChainGPT API key, Chainbase API key); each skill documents its own requirements.
The first-party work in this repository — the master skill and the connectors —
is released under the MIT License. Vendored skills under vendored/
are copies of upstream projects and remain under their own licenses; see each
skill's source repository (and skills-lock.json) for terms.
© 2026 Internet Court Consortium · Open standard, openly governed · internetcourt.org
The living ecosystem where AI agents complete tasks through workflow loops, improve through iterative execution, are evaluated by mentor agents or humans in the loop, and turn completed work into reusable work experience and data to improve future agents.
The living ecosystem where AI agents complete tasks through workflow loops, improve through iterative execution, are evaluated by mentor agents or humans in the loop, and turn completed work into reusable work experience and data to improve future agents.
Agent skills for building, shipping, and growing software products
The self-improving LLM router that optimize your agentic workflows with every runs, works with any harnesses, any models, any loops.
An operating system built for AI agents — talk to your NixOS server instead of SSH-ing in. Typed, audited tool access with atomic rollback on every change. Research-grade; run it on a disposable box, not production.
Vaaya MCP server — pay-per-call agent superpowers: media & video generation, product demo videos, web search & scraping, deep/market research, GTM & sales enrichment, code sandboxes, browser automation, email, memory. No API keys.
The agent-native LLM router for autonomous agents. 55+ models (8 free), <1ms local routing, USDC payments on Base & Solana via x402.
Open Source AI trading agent that operates autonomously across 1000+ markets - Polymarket, Kalshi, Binance, Hyperliquid, Solana DEXs, 5 EVM chains. Scans for edge, executes instantly, manages risk while you sleep. Agent commerce protocol for machine-to-machine payments. Self-hosted. Built on Claude.
Live data for AI agents — search, research, markets, crypto, X/Twitter. Pay-per-call via x402 micropayments.
Open-source MCP servers for Latin American commerce — Pix, NF-e, banking, fiscal, logistics, and messaging across Brazil, Mexico, Argentina, Colombia, Chile, and Peru. MIT, on npm.
Beacon - agent-to-agent pings with optional RTC value attached (BoTTube/Moltbook/RustChain + UDP bus)
Agent skills and commands for Medusa best practices and conventions.