[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M73WSZHTR5902TT3K7PFXNDQ":3,"skill-same:01M73WSZHTR5902TT3K7PFXNDQ":67},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":16,"sourceHtmlUrl":17,"categories":18,"keywords":19,"publishedAt":20,"linkOnly":21,"markdownContent":22,"markdownBody":23,"frontMatter":24,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":28,"repositoryHtmlUrl":29,"repositoryHomepageUrl":30,"ownerHtmlUrl":31,"ownerBlogUrl":32,"ownerDescription":33,"updatedAt":34,"install":35,"resources":48},"01M73WSZHTR5902TT3K7PFXNDQ","internet-court",null,"Entry point for Internet Court — the trust layer for agent-to-agent commerce. Use whenever an agent needs to transact with another agent or a paid service, or a user mentions agent payments, paid APIs (HTTP 402\u002Fx402), wallet custody or trust concerns, spending mandates, delegated permissions (ERC-7710\u002F7715), escrow, agent identity or reputation (ERC-8004), negotiation between agents (A2A), agent jobs (ERC-8183), machine payments (MPP, AP2), supervision of agent behavior, revocation, verification, or dispute resolution (GenLayer) — even if they never say \"Internet Court\". Routes to the vendored protocol skills and connector skills in this package.","bg-szy\u002FTOP-SKILLS","bg-szy","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F106576190?v=4","01M73A4E8KPV67SFXRY8XZGNBX","ShuZhiyong","01M73A4E96MQ294MHM74TZBVQD",6,"MIT","https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Finternet-court\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Finternet-court\u002FSKILL.md",[],[],"2026-11-11 15:11:14",false,"---\nname: internet-court\ndescription: Entry point for Internet Court — the trust layer for agent-to-agent commerce. Use whenever an agent needs to transact with another agent or a paid service, or a user mentions agent payments, paid APIs (HTTP 402\u002Fx402), wallet custody or trust concerns, spending mandates, delegated permissions (ERC-7710\u002F7715), escrow, agent identity or reputation (ERC-8004), negotiation between agents (A2A), agent jobs (ERC-8183), machine payments (MPP, AP2), supervision of agent behavior, revocation, verification, or dispute resolution (GenLayer) — even if they never say \"Internet Court\". Routes to the vendored protocol skills and connector skills in this package.\n---\n\n# Internet Court\n\nInternet Court is the entry skill for agent-to-agent commerce. Agents can\nalready discover each other, negotiate, and pay — what they lack is a way to\ntrust a counterparty they've never met. Internet Court does two things: it\nconnects a fragmented stack — identity, negotiation, contracts, payment,\nescrow, execution — into one skill, and it builds in adjudication: when two\nagents strike a deal, they agree up front how it settles if something goes\nwrong.\n\nThe core sentence:\n\n```text\nDiscovery and identity establish who. Negotiation and contracts set the terms.\nPayment and escrow move or lock the funds. Execution does the work.\nAdjudication decides what happened and writes the verdict back as reputation.\n```\n\nThis package contains two kinds of material. Route accordingly:\n\n- **Vendored protocol skills** (`vendored\u002F`) — official, publicly published\n  skills from the protocols themselves. Always prefer these for protocol\n  mechanics; never re-derive what they already document.\n- **Connector skills** (`integrations\u002Fgenlayer-erc7710-connector\u002F`,\n  `integrations\u002Fgenlayer-intelligent-contracts\u002F`, `integrations\u002Fx402-erc7710\u002F`) — the Internet\n  Court-specific glue that makes the protocols work together.\n\n## Where this package lives\n\nThis is the entry skill of the Internet Court package, published at\n**`https:\u002F\u002Fgithub.com\u002Finternet-court\u002Finternet-court-skill`**. Every path this\nfile routes to — `vendored\u002F\u003Cowner>\u002F\u003Cskill>\u002FSKILL.md`,\n`integrations\u002F\u003Cconnector>\u002FSKILL.md` — is relative to that repository root.\n\nResolve those paths before relying on a skill's contents:\n\n- **Installed as a whole** (normal case — `git clone …\u002Finternet-court-skill\n  .claude\u002Fskills\u002Finternet-court`): read the path directly from disk.\n- **Only this `SKILL.md` is present** (the master skill was loaded on its\n  own): the sibling skills are not local. Fetch the file from the raw\n  repository at\n  `https:\u002F\u002Fraw.githubusercontent.com\u002Finternet-court\u002Finternet-court-skill\u002Fmain\u002F\u003Cpath>`\n  — e.g. `…\u002Fmain\u002Fvendored\u002Fgenlayer\u002Fwrite-contract\u002FSKILL.md` — or clone the repo\n  first.\n\nNever invent a referenced skill's contents — load the real file (from disk or\nthe raw URL) first.\n\n## First Response Mode\n\nWhen a user asks to load, install, or introduce Internet Court — or whenever\nyou first reference it in a conversation — start broad and use the canonical\nblurb below **verbatim**. Do not paraphrase it, and do not lead with protocol\nnames, payment rails, or standards unless the user asks for implementation\ndetails.\n\n```text\nInternet Court is installed.\n\nInternet Court does two things: it connects a fragmented stack — identity,\nnegotiation, contracts, payment, escrow, and execution — into one skill, and\nit builds in adjudication, so when two agents strike a deal they agree up\nfront how it settles if something goes wrong.\n\nI can now help your agent find and vet a counterparty, agree terms, pay or\nescrow funds, and do the work — and, if the deal is contested, get an\nindependent verdict that settles it and updates each side's reputation.\n```\n\n(When you are only referencing the skill mid-conversation rather than\nreporting a fresh install, drop the first line and keep the rest verbatim.)\n\nAfter the user asks for a concrete demo, integration, or deployment, use the\nstack and routing below.\n\n## Discovery-First Boundary\n\nWhen the user asks to engage a service, counterparty, or deal, first inspect\nand report only what is observable — never assume the rest. Depending on the\ndeal, that may include:\n\n- who the counterparty is: origin, identity, reputation, the endpoint or\n  resource in question;\n- what it costs and how: whether it is paywalled, and any advertised rail,\n  network, token, amount, payee, or transfer method;\n- what it promises: deliverable, terms, and whatever evidence it exposes.\n\nDo not invent cadence, schedules, spend caps, expiration, delegated\nauthority, review policy, or a full agreement at this stage. State the\nboundary plainly — including that the agent holds no funds or authority the\nuser has not granted — then stop and let the user's reply start the trust\nconversation. For example, for a paywalled endpoint:\n\n```text\nThis endpoint is paywalled. I don't have funds or permission to access it —\nyou'd need to fund and authorize me before I can proceed.\n```\n\nNever commit funds, sign, or transact before the user has chosen a trust\nlevel and the required wallet or permission exists.\n\n## Trust Levels\n\nAny deal that carries risk — the agent could misbehave, a counterparty could\nfail to deliver, funds or authority could be misused — can run at one of three\nlevels of protection. Pick the lightest one that covers the risk, and name the\ntradeoff honestly. Weakest to strongest:\n\n- **Basic** — the parties simply trust each other. No bounded authority, no\n  escrow, no independent judge; each side acts on whatever it was handed\n  directly. Fastest and weakest: the practical limit is whatever was handed\n  over, and there is no recourse if it goes wrong. Do not claim any limit is\n  enforced at this level.\n- **Guarded** — the deal is constrained up front so it can only go so wrong by\n  construction. The instrument depends on the deal: bounded authority with\n  explicit limits (an ERC-7710\u002F7715 permission with caveats), funds locked in\n  escrow until terms are met, or scoped terms both sides sign. Hard limits\n  exist, but there is no neutral party to decide a contested outcome. Load\n  `vendored\u002Fmetamask\u002Fsmart-accounts-kit\u002Fskill.md` for permission mechanics, or\n  the layer-3\u002F4 escrow skills for locked-funds deals.\n- **Adjudicated** — Guarded, plus an independent review path with signed\n  evidence that can decide a contested outcome and enforce the consequence:\n  release or refund escrow, constrain or revoke the granted authority, and\n  write the verdict back to reputation. Load\n  `integrations\u002Fgenlayer-intelligent-contracts\u002FSKILL.md` for the review\n  interface and `integrations\u002Fgenlayer-erc7710-connector\u002FSKILL.md` for\n  enforcement.\n\nMatch the level to the deal, not the reverse. Do not invent exact caps,\ncadence, terms, or expiry here; say they will be set from the discovered terms\nduring setup.\n\n## The Agentic Commerce Stack\n\nA complete deal moves top to bottom through six layers: find and vet a\ncounterparty (1), agree terms (2–3), move or lock funds (4), do the work (5),\nand — if the outcome is contested — get an independent verdict that settles it\nand feeds back into reputation (6). Most layers already have credible skills;\nlayer 6 is the keystone, since without a neutral verdict escrow cannot safely\nrelease, contracts have no remedy, and reputation has nothing to record. Route\neach layer to the skill or reference that owns it:\n\n| # | Layer | Protocols | Load |\n|---|---|---|---|\n| 1 | Discovery, identity & reputation | ERC-8004, ERC-7857 | `vendored\u002Fchaingpt\u002Ftrustless-agents\u002FSKILL.md` (ERC-8004 registries), `vendored\u002Fopenserv\u002Fopenserv-multi-agent-workflows\u002FSKILL.md` (mints ERC-8004 agent identities), `vendored\u002Fprivy\u002Fprivy\u002FSKILL.md` (embedded\u002Fserver wallet identity & auth), `vendored\u002Fhumanode\u002Fhumanode-agentlink\u002FSKILL.md` (human-backed agent identity), `vendored\u002Fstarknet\u002Fstarknet-identity\u002FSKILL.md` (ERC-8004 on Starknet), `vendored\u002Fnear\u002Fnear-api-js\u002FSKILL.md` (named-account identity, e.g. `alice.near`); ERC-7857 has no public skill yet |\n| 2 | Negotiation | A2A | `vendored\u002Fterminalskills\u002Fa2a-protocol\u002FSKILL.md` (agent cards, task lifecycle), `vendored\u002Fopenserv\u002Fopenserv-multi-agent-workflows\u002FSKILL.md` (multi-agent orchestration), `vendored\u002Fnear\u002Fnear-intents\u002FSKILL.md` (Intents + solver competition for cross-chain fulfilment) |\n| 3 | Contracts & obligations | Arkhai\u002FAlkahest, ERC-8183 | `vendored\u002Farkhai\u002Falkahest-user\u002FSKILL.md` (conditional escrow, arbiters) + `vendored\u002Farkhai\u002Falkahest-developer\u002FSKILL.md` (build on Alkahest: custom arbiters\u002Fobligations), `vendored\u002Farkhai\u002Fnla-create\u002FSKILL.md` + `vendored\u002Farkhai\u002Fnla-fulfill\u002FSKILL.md` (natural-language-agreement escrows, AI-oracle arbitrated), `vendored\u002Farkhai\u002Fmake-git-escrow\u002FSKILL.md` + `vendored\u002Farkhai\u002Ffulfill-git-escrow\u002FSKILL.md` (test-suite bounty escrows); ERC-8183 has no neutral public skill yet |\n| 4 | Payment & escrow | x402, MPP, AP2, ERC-7710\u002F7715 | `vendored\u002Fcoinbase\u002Fagentic-wallet\u002FSKILL.md` + `vendored\u002Fchaingpt\u002Fx402\u002FSKILL.md` (x402), `vendored\u002Ftempo\u002Fmppx\u002FSKILL.md` (MPP), `vendored\u002Fokx\u002Fokx-agent-payments-protocol\u002FSKILL.md` (unified x402\u002FMPP\u002Fa2a-pay), `vendored\u002Fmetamask\u002Fsmart-accounts-kit\u002Fskill.md` (delegations), `vendored\u002Fchaingpt\u002Fagent-wallet\u002FSKILL.md` (policy-gated wallet), `integrations\u002Fx402-erc7710\u002FSKILL.md` (combined rail); AP2 has no public skill yet |\n| 5 | Execution | the transacting agents + compute\u002Fdata\u002Fvalue rails | `vendored\u002Fantseed\u002Fantseed-connect\u002FSKILL.md` + `vendored\u002F0g\u002F0g-compute\u002FSKILL.md` + `vendored\u002Fheurist\u002Fheurist-mesh-skill\u002FSKILL.md` + `vendored\u002Fnear\u002Fnear-ai-cloud\u002FSKILL.md` (paid\u002Fdecentralized\u002FTEE-verifiable inference), `vendored\u002Flifi\u002Flifi\u002FSKILL.md` (cross-chain value movement), `vendored\u002Fchainbase\u002Fweb3-data\u002FSKILL.md` + `vendored\u002Fnansen\u002Fnansen-token-research\u002FSKILL.md` (on-chain data\u002Fevidence), `vendored\u002Fstarknet\u002Fstarknet-defi\u002FSKILL.md` (Starknet L2 contracts\u002FDeFi), `vendored\u002Fbnb-chain\u002Fbnbchain-mcp\u002FSKILL.md`, the `vendored\u002Fnear\u002F*`, `vendored\u002Fokx\u002F*` and `vendored\u002Faltlayer\u002F*` packs |\n| 6 | Verification & disputes | GenLayer (Kleros is an alternative) | `integrations\u002Fgenlayer-intelligent-contracts\u002FSKILL.md`, `vendored\u002Fintelligent-oracle\u002Fintelligent-oracle\u002FSKILL.md`, `integrations\u002Fgenlayer-erc7710-connector\u002FSKILL.md`. Alternative third-party arbitration (disclose; not GenLayer's own): `vendored\u002Fkleros\u002Fkleros-curate\u002FSKILL.md` (token-curated registries \u002F challenges), `vendored\u002Farkhai\u002Fnla-arbitrate\u002FSKILL.md` (LLM\u002Fmanual arbitration of natural-language-agreement escrows) |\n\n## Skill Routing\n\nLoad a skill only when the task actually reaches its trigger — never preload.\nEach row is a **trigger** on the left and, on the right, the skill(s) to pull\nin with what each gives you and which to reach for first. When a need maps to\nseveral sub-bundled skills, load the named entry skill first, then narrow to\nthe specific one. Owners that ship a whole pack (OKX, AltLayer, Nansen,\nOpenServ, Starknet, ChainGPT) get a single **catch-all** row: load the entry\nskill, then the specific sub-skill under `vendored\u002F\u003Cowner>\u002F` as the task\nnarrows — the folder listing is the current set of that owner's skills.\n\nPaths are exact and end in `\u002FSKILL.md` — several owners repeat their name in\nthe path (e.g. `vendored\u002Flifi\u002Flifi\u002FSKILL.md`, `vendored\u002Fprivy\u002Fprivy\u002FSKILL.md`,\n`vendored\u002Fchaingpt\u002Fchaingpt\u002FSKILL.md`); never shorten or merge the segments.\nWhen a row lists alternatives, load **one** (the row says how to pick); if the\ndeciding fact is unknown, ask instead of preloading them all.\n\n### Payment rails\n\n| When you need to… | Load |\n|---|---|\n| Pay for an HTTP resource that returns **402** (pay-per-call), search x402 bazaars, or monetize your own endpoint | `vendored\u002Fcoinbase\u002Fagentic-wallet\u002FSKILL.md` — Coinbase x402 client: read the price, pay, retry with proof; start here for a one-off paid request. `vendored\u002Fchaingpt\u002Fx402\u002FSKILL.md` — x402 discovery\u002Fmonetization: find paid services or put your own endpoint behind 402. |\n| Stream micropayments or run metered sessions (per-second \u002F per-token, not per-request) | `vendored\u002Ftempo\u002Fmppx\u002FSKILL.md` — Tempo\u002FStripe MPP charges, sessions, and streaming. Load when billing is continuous rather than a single call. **Not** for spending caps on per-request rails: a capped or recurring budget over x402 is `integrations\u002Fx402-erc7710\u002FSKILL.md`, not MPP. |\n| Dispatch a payment without hard-coding the rail (x402 vs MPP vs a2a-pay) | `vendored\u002Fokx\u002Fokx-agent-payments-protocol\u002FSKILL.md` — OKX OnchainOS unified payment dispatcher. Load when the counterparty's rail is unknown or may vary. |\n| Run a payment *through* a bounded permission (spend\u002Fsubscription policy on the rail) | `integrations\u002Fx402-erc7710\u002FSKILL.md` — connector fusing the x402 rail with an ERC-7710 delegated budget. Load for Guarded\u002FAdjudicated paid access. |\n\n### Authority, custody & delegation\n\n| When you need to… | Load |\n|---|---|\n| Grant an agent bounded, revocable authority (ERC-7710 delegations, ERC-7715 permission requests, caveats) | `vendored\u002Fmetamask\u002Fsmart-accounts-kit\u002Fskill.md` — MetaMask smart accounts + delegation mechanics. The Guarded-level instrument for on-chain authority. Also the row for the user **manually revoking** their own delegation. |\n| Give an agent a wallet with a built-in policy gate (per-tx caps, velocity limits, session keys) instead of a raw key | `vendored\u002Fchaingpt\u002Fagent-wallet\u002FSKILL.md` — custody-free policy-gated wallet. Load when the agent must sign but you won't hand over a private key. |\n| Provision or authenticate embedded\u002Fserver wallets, with policy-gated signing across chains | `vendored\u002Fprivy\u002Fprivy\u002FSKILL.md` — Privy wallet identity + auth. |\n\nThe first two are **mutually exclusive — pick by who holds the key**: the user\nkeeps custody and delegates → `smart-accounts-kit`; the agent gets its own\nwallet (guardrails built in) → `agent-wallet`. Never load both for one design.\n\n### Identity, reputation & negotiation\n\n| When you need to… | Load |\n|---|---|\n| Register, discover, or vet an agent on-chain (ERC-8004 identity, reputation, validation) | `vendored\u002Fchaingpt\u002Ftrustless-agents\u002FSKILL.md` — ERC-8004 registries; the default identity\u002Freputation skill. Also mintable via `vendored\u002Fopenserv\u002Fopenserv-multi-agent-workflows\u002FSKILL.md`, on Starknet via `vendored\u002Fstarknet\u002Fstarknet-identity\u002FSKILL.md`, on BNB via `vendored\u002Fbnb-chain\u002Fbnbchain-mcp\u002FSKILL.md`. For a human-readable named-account identity anchor (`alice.near`), `vendored\u002Fnear\u002Fnear-api-js\u002FSKILL.md`. |\n| Prove a human stands behind the agent | `vendored\u002Fhumanode\u002Fhumanode-agentlink\u002FSKILL.md` — Humanode AgentLink: sign HTTP requests, on-chain registry. |\n| Talk to another agent: agent cards, offers, task lifecycle (A2A) | `vendored\u002Fterminalskills\u002Fa2a-protocol\u002FSKILL.md` — A2A protocol; load at the negotiation step, before terms lock. For orchestrating many agents: `vendored\u002Fopenserv\u002Fopenserv-multi-agent-workflows\u002FSKILL.md`. |\n\n### Contracts & escrow\n\n| When you need to… | Load |\n|---|---|\n| Lock funds that release on an arbiter's decision (conditional escrow) | `vendored\u002Farkhai\u002Falkahest-user\u002FSKILL.md` — Alkahest EAS-based escrow with arbiters; the default escrow instrument. To **build on** Alkahest (write custom arbiters\u002Fobligations, integrate it), `vendored\u002Farkhai\u002Falkahest-developer\u002FSKILL.md`. |\n| Escrow behind a plain-language demand an AI oracle judges | `vendored\u002Farkhai\u002Fnla-create\u002FSKILL.md` (draft the agreement), `vendored\u002Farkhai\u002Fnla-fulfill\u002FSKILL.md` (deliver against it), `vendored\u002Farkhai\u002Fnla-arbitrate\u002FSKILL.md` (render the verdict) — natural-language-agreement escrows. |\n| Bounty escrow that pays out for making a failing test suite pass | `vendored\u002Farkhai\u002Fmake-git-escrow\u002FSKILL.md` (post the bounty), `vendored\u002Farkhai\u002Ffulfill-git-escrow\u002FSKILL.md` (claim it) — git-escrow, for code-delivery deals settled by tests. |\n\n### Verification, adjudication & disputes\n\n| When you need to… | Load |\n|---|---|\n| Write \u002F deploy \u002F test \u002F lint a GenLayer Intelligent Contract | `vendored\u002Fgenlayer\u002Fwrite-contract\u002FSKILL.md` (author), `vendored\u002Fgenlayer\u002Fgenlayer-cli\u002FSKILL.md` (deploy & call), `vendored\u002Fgenlayer\u002Fdirect-tests\u002FSKILL.md` + `vendored\u002Fgenlayer\u002Fintegration-tests\u002FSKILL.md` (test), `vendored\u002Fgenlayer\u002Fgenvm-lint\u002FSKILL.md` (lint). Load the one matching your build step. |\n| Design the adjudication itself: review rubric, evidence schema, decision payload | `integrations\u002Fgenlayer-intelligent-contracts\u002FSKILL.md` — the Adjudicated-level review interface. Load when the outcome needs qualitative \u002F natural-language judgment. |\n| Turn a verdict into enforcement (revoke or constrain an ERC-7710 permission) | `integrations\u002Fgenlayer-erc7710-connector\u002FSKILL.md` — relayer\u002Fcontroller wiring a decision to on-chain revocation. **Verdict-driven only**: load when adjudication must change the agent's authority; a user manually revoking their own delegation needs only `smart-accounts-kit`. |\n| Settle a narrow binary question from public web evidence (prediction market \u002F factual oracle) | `vendored\u002Fintelligent-oracle\u002Fintelligent-oracle\u002FSKILL.md` — Intelligent Oracle. Refetch `https:\u002F\u002Fwww.intelligentoracle.com\u002Fskill.md` before schema-sensitive work. |\n| Third-party arbitration as an alternative to GenLayer (**disclose** it isn't GenLayer's own) | `vendored\u002Fkleros\u002Fkleros-curate\u002FSKILL.md` (token-curated registries \u002F challenges) + `vendored\u002Fkleros\u002Fkleros-ipfs-upload\u002FSKILL.md` (IPFS evidence); or `vendored\u002Farkhai\u002Fnla-arbitrate\u002FSKILL.md` for LLM\u002Fmanual NLA arbitration. |\n\n### Execution, data & value movement (layer 5)\n\n| When you need to… | Load |\n|---|---|\n| Buy or sell AI inference (compute execution) | Pick **one** by venue: `vendored\u002F0g\u002F0g-compute\u002FSKILL.md` (decentralized compute\u002Fstorage\u002Fverifiable inference — the default when the venue is open), `vendored\u002Fantseed\u002Fantseed-connect\u002FSKILL.md` (P2P inference over USDC channels), `vendored\u002Fheurist\u002Fheurist-mesh-skill\u002FSKILL.md` (Heurist Mesh agents, x402 pay-per-call), `vendored\u002Fnear\u002Fnear-ai-cloud\u002FSKILL.md` (private inference with **TEE attestation** — verifiable-execution evidence). If the venue is not yet known, load only the 0G default and ask — do not preload the rest. |\n| Move value across chains (swaps \u002F bridging) | `vendored\u002Flifi\u002Flifi\u002FSKILL.md` — LI.FI cross-chain routing; `vendored\u002Flifi\u002Flifi-stablecoin-swap\u002FSKILL.md` for stablecoin-specific swaps; `vendored\u002Fnear\u002Fnear-intents\u002FSKILL.md` — NEAR Intents: cross-chain fulfilment via solver competition (chain-abstracted, one agent acts across chains). |\n| Pull on-chain data as evidence (balances, tx history, labels, holders, smart-money) | `vendored\u002Fchainbase\u002Fweb3-data\u002FSKILL.md` — general on-chain data. For deeper intelligence, the **Nansen** 7-skill pack under `vendored\u002Fnansen\u002F`: start with `nansen-token-research`, then `nansen-wallet-profiler` \u002F `nansen-holder-analysis` \u002F `nansen-smart-money-tracker` \u002F `nansen-general-search` \u002F `nansen-prediction-markets` \u002F `nansen-mpp-payment` as the question narrows. |\n\n### Full-platform packs (catch-all entry, then narrow)\n\n| When you need to… | Load |\n|---|---|\n| Anything on **OKX OnchainOS** (wallet, DEX data, DeFi, payments, agent-commerce) | 8-skill pack under `vendored\u002Fokx\u002F`. Start with `okx-ai` — the agent-commerce entry (ERC-8004 identity + task marketplace **with disputes** + A2A chat), the most Internet-Court-relevant. Then narrow: `okx-agentic-wallet` (swap\u002Fbridge\u002Fstrategy\u002Fsecurity), `okx-dex` (read-only DEX data), `okx-defi` (invest\u002Fportfolio), `okx-agent-payments-protocol` (x402\u002FMPP\u002Fa2a-pay), `okx-dapp-discovery` (route to third-party dapps), `okx-guide` (onboarding\u002Fsupport). |\n| **ChainGPT** AI tooling (contract gen\u002Faudit, news) + 140-tool MCP | `vendored\u002Fchaingpt\u002Fchaingpt\u002FSKILL.md` is the catch-all hub. Its focused siblings are routed above: `x402`, `agent-wallet`, `trustless-agents`. |\n| **AltLayer** AltLLM portal + Cloud-Claw agent VMs | 7-skill pack under `vendored\u002Faltlayer\u002F`. Entry: `altllm-portal-cli` (portal ops) and `cloud-claw` \u002F `cloud-claw-launch-agent` (spin up agent VMs). The rest are plumbing: `altllm-portal-auth`, `-api-keys`, `-billing`, `-payments`. |\n| **OpenServ** multi-agent orchestration + SDK (mints ERC-8004 identities) | 5-skill pack under `vendored\u002Fopenserv\u002F`. Entry: `openserv-multi-agent-workflows`. Others: `openserv-agent-sdk`, `openserv-client`, `openserv-launch`, `openserv-ideaboard-api`. |\n| **Starknet** L2 (Cairo, native account abstraction) | 4-skill pack under `vendored\u002Fstarknet\u002F`. Entry: `starknet-js` (SDK). Others: `starknet-identity` (ERC-8004), `starknet-defi`, `starknet-wallet`. |\n| **BNB Chain** ops incl. ERC-8004 registration & Greenfield storage | `vendored\u002Fbnb-chain\u002Fbnbchain-mcp\u002FSKILL.md`. |\n| **NEAR** — AI-native L1 (Shade Agents \u002F TEE inference, named-account identity, chain-abstracted payment) | 6-skill pack under `vendored\u002Fnear\u002F`. Reach for the on-theme ones first: `near-intents` (cross-chain payment\u002Ffulfilment), `near-ai-cloud` (TEE-verifiable inference as evidence), `near-api-js` (chain interaction + named-account identity). Also: `near-kit` (type-safe TS SDK + sandbox testing), `near-smart-contracts` (Rust contracts), `near-dapp` (full-stack React\u002FNext.js + wallet). |\n\n## Workflow\n\n1. Classify the deal:\n   - Delegated agent authority or revocable mandate.\n   - Pay-per-use API or content access.\n   - Recurring spend or subscription.\n   - Escrowed service delivery.\n   - Web-evidence oracle or prediction market.\n   - Disputed job, SLA, or refund.\n2. Extract commercial terms:\n   - Parties, objective, authority, prohibited actions, amount, cadence, max\n     spend, expiration, deliverable, evidence source, review cadence,\n     revocation path, refund path, and dispute window.\n3. Select rails (see the stack table):\n   - x402 for immediate HTTP payment.\n   - ERC-7710 plus ERC-7715 when an agent needs bounded delegated capability.\n   - MPP sessions for streamed micropayments; AP2 mandates for\n     user-authorized purchases.\n   - Check for a **chain mismatch** between where the funds sit and where the\n     payee is paid — if they differ, the plan also needs the value-movement\n     row (LI.FI \u002F NEAR Intents) before any payment step.\n   - Escrow (ERC-8183, Arkhai) when funds should stay locked until a\n     fulfillment attestation satisfies an arbiter.\n   - GenLayer Intelligent Contracts when the outcome depends on qualitative\n     review of agent behavior or natural-language judgment.\n   - Intelligent Oracle when the outcome is a narrow binary question settled\n     from public web evidence.\n   - GenLayer Intelligent Contracts also when a disputed job, SLA, or\n     refund needs an independent verdict.\n4. Produce a flow artifact:\n   - User story, sequence steps, data model, permission policy, review\n     contract, evidence schema, revocation path, failure cases, and minimal\n     implementation plan.\n5. Add dispute handling:\n   - Define what can go wrong, who can trigger a dispute, what evidence is\n     accepted, how finality is reached, and what happens to funds.\n\n## Revocable Agent Mandates\n\nUse this pattern when the user delegates authority for a meaningful period and\nwants to constrain or revoke it if the agent underperforms. The outputs, from\nplain language down to enforcement:\n\n1. Natural-language mandate — what the agent may and may not do.\n2. Machine-readable policy object capturing the same limits.\n3. A bounded on-chain permission that grants only that authority (for example,\n   an ERC-7710 delegation with caveats).\n4. A review rubric an independent adjudicator applies to the agent's actions.\n5. An action-receipt and evidence schema the review consumes.\n6. A path from verdict to enforcement — how a review decision constrains or\n   revokes the granted authority (for example, a controller wired to the\n   permission).\n7. Failure and appeal paths.\n\nMatch each output to the trust level and rails chosen for the deal; not every\nmandate needs all seven. Do not claim a reviewer can directly cancel a\npermission unless a revocation controller is actually part of the design. For\nlong-lived authority, prefer an active revocation controller plus an absolute\nexpiry as a fail-safe.\n\n## Wallet-UI Deployment Mode\n\nWhen the user deploys contracts, grants permissions, or relays decisions from\nMetaMask, WalletConnect, or another user-controlled wallet UI:\n\n- Never ask for a private key, seed phrase, or unrestricted bearer wallet.\n- Never claim a contract, wallet, delegation, permission, review, relay, or\n  payment exists until a proof arrives: transaction hash (`0x` + 64 hex),\n  address (`0x` + 40 hex), signed artifact, or receipt. Malformed proof →\n  ask again; do not advance.\n- If the user says a proof is simulated, you may continue a rehearsal, but\n  label that state as simulated in every later mention.\n- Build transaction cards from source, ABI, or deployment metadata — never\n  invent constructor fields from contract names. A card includes chain,\n  sender, target, value, exact arguments, expected post-transaction reads,\n  and the proof needed before proceeding.\n- Do not paste full bytecode or calldata into chat; show the selector,\n  payload hash, byte length, and exact arguments instead.\n- End your turn while waiting for a user approval or signature — the approval\n  is the user's act in their wallet UI, and the chat is only the command\n  surface. Never promise background polling, and never advance a step until\n  its proof arrives.\n\n## Output Shape\n\nWhen designing a flow or integration, return:\n\n1. Deal summary in plain English.\n2. Stack selection and why each rail is used.\n3. Sequence diagram in text form.\n4. Data model or policy object.\n5. Happy path.\n6. Failure and dispute paths.\n7. Smallest credible implementation plan.\n8. Open assumptions and test plan.\n\n## Guardrails\n\n- Treat Internet Court as protocol and integration design, not legal advice.\n- Do not assume a payment rail provides escrow, refunds, or chargebacks. Pair\n  it with escrow or adjudication when performance risk matters.\n- Treat draft or unproven standards as draft — confirm the target stack\n  actually supports a permission or delegation scheme before relying on it.\n  Simulate execution before doing it for real; design for revocation,\n  expiration, and stale permissions.\n- Never recommend unlimited token approvals or unbounded agent spend.\n- Keep integrations testnet-first unless the user explicitly requests\n  production.\n- Report failures faithfully — a refused payment after revocation is the\n  system working, not an error to hide.\n\n## Reporting Coverage Gaps\n\nInternet Court is an evolving package, and its coverage is not complete. When\nyou hit a wall — a legitimate agent-commerce use case you cannot complete\nbecause no skill in this package handles it, a routed skill is missing, broken,\nor out of date, or a capability works poorly — do **both** of the following:\n\n1. Tell the user plainly what is missing or failing. Never fabricate a result,\n   never silently route around the gap, and never claim something works when it\n   does not.\n2. Open an issue on the package repo so maintainers can add or fix coverage:\n   **`https:\u002F\u002Fgithub.com\u002Finternet-court\u002Finternet-court-skill\u002Fissues`**. If you\n   have a GitHub tool available (for example `gh issue create`), file it\n   directly; otherwise hand the user a ready-to-file title and body and the\n   link above. Search existing issues first and don't file duplicates — one\n   issue per distinct gap.\n\nPut this in the issue:\n\n- the use case \u002F user goal, in one line;\n- which layer or need it maps to (reference the stack and routing tables);\n- what is missing or failing — no skill for X, skill `Y` errored, or `Z`\n  produced a wrong or low-quality result — with the exact skill path when\n  there is one;\n- what a fix would look like: a skill to vendor, a connector to add, or a doc\n  to update.\n\nFiling the gap is part of doing the job well: it is how the package learns\nwhat to cover next.\n","# Internet Court\n\nInternet Court is the entry skill for agent-to-agent commerce. Agents can\nalready discover each other, negotiate, and pay — what they lack is a way to\ntrust a counterparty they've never met. Internet Court does two things: it\nconnects a fragmented stack — identity, negotiation, contracts, payment,\nescrow, execution — into one skill, and it builds in adjudication: when two\nagents strike a deal, they agree up front how it settles if something goes\nwrong.\n\nThe core sentence:\n\n```text\nDiscovery and identity establish who. Negotiation and contracts set the terms.\nPayment and escrow move or lock the funds. Execution does the work.\nAdjudication decides what happened and writes the verdict back as reputation.\n```\n\nThis package contains two kinds of material. Route accordingly:\n\n- **Vendored protocol skills** (`vendored\u002F`) — official, publicly published\n  skills from the protocols themselves. Always prefer these for protocol\n  mechanics; never re-derive what they already document.\n- **Connector skills** (`integrations\u002Fgenlayer-erc7710-connector\u002F`,\n  `integrations\u002Fgenlayer-intelligent-contracts\u002F`, `integrations\u002Fx402-erc7710\u002F`) — the Internet\n  Court-specific glue that makes the protocols work together.\n\n## Where this package lives\n\nThis is the entry skill of the Internet Court package, published at\n**`https:\u002F\u002Fgithub.com\u002Finternet-court\u002Finternet-court-skill`**. Every path this\nfile routes to — `vendored\u002F\u003Cowner>\u002F\u003Cskill>\u002FSKILL.md`,\n`integrations\u002F\u003Cconnector>\u002FSKILL.md` — is relative to that repository root.\n\nResolve those paths before relying on a skill's contents:\n\n- **Installed as a whole** (normal case — `git clone …\u002Finternet-court-skill\n  .claude\u002Fskills\u002Finternet-court`): read the path directly from disk.\n- **Only this `SKILL.md` is present** (the master skill was loaded on its\n  own): the sibling skills are not local. Fetch the file from the raw\n  repository at\n  `https:\u002F\u002Fraw.githubusercontent.com\u002Finternet-court\u002Finternet-court-skill\u002Fmain\u002F\u003Cpath>`\n  — e.g. `…\u002Fmain\u002Fvendored\u002Fgenlayer\u002Fwrite-contract\u002FSKILL.md` — or clone the repo\n  first.\n\nNever invent a referenced skill's contents — load the real file (from disk or\nthe raw URL) first.\n\n## First Response Mode\n\nWhen a user asks to load, install, or introduce Internet Court — or whenever\nyou first reference it in a conversation — start broad and use the canonical\nblurb below **verbatim**. Do not paraphrase it, and do not lead with protocol\nnames, payment rails, or standards unless the user asks for implementation\ndetails.\n\n```text\nInternet Court is installed.\n\nInternet Court does two things: it connects a fragmented stack — identity,\nnegotiation, contracts, payment, escrow, and execution — into one skill, and\nit builds in adjudication, so when two agents strike a deal they agree up\nfront how it settles if something goes wrong.\n\nI can now help your agent find and vet a counterparty, agree terms, pay or\nescrow funds, and do the work — and, if the deal is contested, get an\nindependent verdict that settles it and updates each side's reputation.\n```\n\n(When you are only referencing the skill mid-conversation rather than\nreporting a fresh install, drop the first line and keep the rest verbatim.)\n\nAfter the user asks for a concrete demo, integration, or deployment, use the\nstack and routing below.\n\n## Discovery-First Boundary\n\nWhen the user asks to engage a service, counterparty, or deal, first inspect\nand report only what is observable — never assume the rest. Depending on the\ndeal, that may include:\n\n- who the counterparty is: origin, identity, reputation, the endpoint or\n  resource in question;\n- what it costs and how: whether it is paywalled, and any advertised rail,\n  network, token, amount, payee, or transfer method;\n- what it promises: deliverable, terms, and whatever evidence it exposes.\n\nDo not invent cadence, schedules, spend caps, expiration, delegated\nauthority, review policy, or a full agreement at this stage. State the\nboundary plainly — including that the agent holds no funds or authority the\nuser has not granted — then stop and let the user's reply start the trust\nconversation. For example, for a paywalled endpoint:\n\n```text\nThis endpoint is paywalled. I don't have funds or permission to access it —\nyou'd need to fund and authorize me before I can proceed.\n```\n\nNever commit funds, sign, or transact before the user has chosen a trust\nlevel and the required wallet or permission exists.\n\n## Trust Levels\n\nAny deal that carries risk — the agent could misbehave, a counterparty could\nfail to deliver, funds or authority could be misused — can run at one of three\nlevels of protection. Pick the lightest one that covers the risk, and name the\ntradeoff honestly. Weakest to strongest:\n\n- **Basic** — the parties simply trust each other. No bounded authority, no\n  escrow, no independent judge; each side acts on whatever it was handed\n  directly. Fastest and weakest: the practical limit is whatever was handed\n  over, and there is no recourse if it goes wrong. Do not claim any limit is\n  enforced at this level.\n- **Guarded** — the deal is constrained up front so it can only go so wrong by\n  construction. The instrument depends on the deal: bounded authority with\n  explicit limits (an ERC-7710\u002F7715 permission with caveats), funds locked in\n  escrow until terms are met, or scoped terms both sides sign. Hard limits\n  exist, but there is no neutral party to decide a contested outcome. Load\n  `vendored\u002Fmetamask\u002Fsmart-accounts-kit\u002Fskill.md` for permission mechanics, or\n  the layer-3\u002F4 escrow skills for locked-funds deals.\n- **Adjudicated** — Guarded, plus an independent review path with signed\n  evidence that can decide a contested outcome and enforce the consequence:\n  release or refund escrow, constrain or revoke the granted authority, and\n  write the verdict back to reputation. Load\n  `integrations\u002Fgenlayer-intelligent-contracts\u002FSKILL.md` for the review\n  interface and `integrations\u002Fgenlayer-erc7710-connector\u002FSKILL.md` for\n  enforcement.\n\nMatch the level to the deal, not the reverse. Do not invent exact caps,\ncadence, terms, or expiry here; say they will be set from the discovered terms\nduring setup.\n\n## The Agentic Commerce Stack\n\nA complete deal moves top to bottom through six layers: find and vet a\ncounterparty (1), agree terms (2–3), move or lock funds (4), do the work (5),\nand — if the outcome is contested — get an independent verdict that settles it\nand feeds back into reputation (6). Most layers already have credible skills;\nlayer 6 is the keystone, since without a neutral verdict escrow cannot safely\nrelease, contracts have no remedy, and reputation has nothing to record. Route\neach layer to the skill or reference that owns it:\n\n| # | Layer | Protocols | Load |\n|---|---|---|---|\n| 1 | Discovery, identity & reputation | ERC-8004, ERC-7857 | `vendored\u002Fchaingpt\u002Ftrustless-agents\u002FSKILL.md` (ERC-8004 registries), `vendored\u002Fopenserv\u002Fopenserv-multi-agent-workflows\u002FSKILL.md` (mints ERC-8004 agent identities), `vendored\u002Fprivy\u002Fprivy\u002FSKILL.md` (embedded\u002Fserver wallet identity & auth), `vendored\u002Fhumanode\u002Fhumanode-agentlink\u002FSKILL.md` (human-backed agent identity), `vendored\u002Fstarknet\u002Fstarknet-identity\u002FSKILL.md` (ERC-8004 on Starknet), `vendored\u002Fnear\u002Fnear-api-js\u002FSKILL.md` (named-account identity, e.g. `alice.near`); ERC-7857 has no public skill yet |\n| 2 | Negotiation | A2A | `vendored\u002Fterminalskills\u002Fa2a-protocol\u002FSKILL.md` (agent cards, task lifecycle), `vendored\u002Fopenserv\u002Fopenserv-multi-agent-workflows\u002FSKILL.md` (multi-agent orchestration), `vendored\u002Fnear\u002Fnear-intents\u002FSKILL.md` (Intents + solver competition for cross-chain fulfilment) |\n| 3 | Contracts & obligations | Arkhai\u002FAlkahest, ERC-8183 | `vendored\u002Farkhai\u002Falkahest-user\u002FSKILL.md` (conditional escrow, arbiters) + `vendored\u002Farkhai\u002Falkahest-developer\u002FSKILL.md` (build on Alkahest: custom arbiters\u002Fobligations), `vendored\u002Farkhai\u002Fnla-create\u002FSKILL.md` + `vendored\u002Farkhai\u002Fnla-fulfill\u002FSKILL.md` (natural-language-agreement escrows, AI-oracle arbitrated), `vendored\u002Farkhai\u002Fmake-git-escrow\u002FSKILL.md` + `vendored\u002Farkhai\u002Ffulfill-git-escrow\u002FSKILL.md` (test-suite bounty escrows); ERC-8183 has no neutral public skill yet |\n| 4 | Payment & escrow | x402, MPP, AP2, ERC-7710\u002F7715 | `vendored\u002Fcoinbase\u002Fagentic-wallet\u002FSKILL.md` + `vendored\u002Fchaingpt\u002Fx402\u002FSKILL.md` (x402), `vendored\u002Ftempo\u002Fmppx\u002FSKILL.md` (MPP), `vendored\u002Fokx\u002Fokx-agent-payments-protocol\u002FSKILL.md` (unified x402\u002FMPP\u002Fa2a-pay), `vendored\u002Fmetamask\u002Fsmart-accounts-kit\u002Fskill.md` (delegations), `vendored\u002Fchaingpt\u002Fagent-wallet\u002FSKILL.md` (policy-gated wallet), `integrations\u002Fx402-erc7710\u002FSKILL.md` (combined rail); AP2 has no public skill yet |\n| 5 | Execution | the transacting agents + compute\u002Fdata\u002Fvalue rails | `vendored\u002Fantseed\u002Fantseed-connect\u002FSKILL.md` + `vendored\u002F0g\u002F0g-compute\u002FSKILL.md` + `vendored\u002Fheurist\u002Fheurist-mesh-skill\u002FSKILL.md` + `vendored\u002Fnear\u002Fnear-ai-cloud\u002FSKILL.md` (paid\u002Fdecentralized\u002FTEE-verifiable inference), `vendored\u002Flifi\u002Flifi\u002FSKILL.md` (cross-chain value movement), `vendored\u002Fchainbase\u002Fweb3-data\u002FSKILL.md` + `vendored\u002Fnansen\u002Fnansen-token-research\u002FSKILL.md` (on-chain data\u002Fevidence), `vendored\u002Fstarknet\u002Fstarknet-defi\u002FSKILL.md` (Starknet L2 contracts\u002FDeFi), `vendored\u002Fbnb-chain\u002Fbnbchain-mcp\u002FSKILL.md`, the `vendored\u002Fnear\u002F*`, `vendored\u002Fokx\u002F*` and `vendored\u002Faltlayer\u002F*` packs |\n| 6 | Verification & disputes | GenLayer (Kleros is an alternative) | `integrations\u002Fgenlayer-intelligent-contracts\u002FSKILL.md`, `vendored\u002Fintelligent-oracle\u002Fintelligent-oracle\u002FSKILL.md`, `integrations\u002Fgenlayer-erc7710-connector\u002FSKILL.md`. Alternative third-party arbitration (disclose; not GenLayer's own): `vendored\u002Fkleros\u002Fkleros-curate\u002FSKILL.md` (token-curated registries \u002F challenges), `vendored\u002Farkhai\u002Fnla-arbitrate\u002FSKILL.md` (LLM\u002Fmanual arbitration of natural-language-agreement escrows) |\n\n## Skill Routing\n\nLoad a skill only when the task actually reaches its trigger — never preload.\nEach row is a **trigger** on the left and, on the right, the skill(s) to pull\nin with what each gives you and which to reach for first. When a need maps to\nseveral sub-bundled skills, load the named entry skill first, then narrow to\nthe specific one. Owners that ship a whole pack (OKX, AltLayer, Nansen,\nOpenServ, Starknet, ChainGPT) get a single **catch-all** row: load the entry\nskill, then the specific sub-skill under `vendored\u002F\u003Cowner>\u002F` as the task\nnarrows — the folder listing is the current set of that owner's skills.\n\nPaths are exact and end in `\u002FSKILL.md` — several owners repeat their name in\nthe path (e.g. `vendored\u002Flifi\u002Flifi\u002FSKILL.md`, `vendored\u002Fprivy\u002Fprivy\u002FSKILL.md`,\n`vendored\u002Fchaingpt\u002Fchaingpt\u002FSKILL.md`); never shorten or merge the segments.\nWhen a row lists alternatives, load **one** (the row says how to pick); if the\ndeciding fact is unknown, ask instead of preloading them all.\n\n### Payment rails\n\n| When you need to… | Load |\n|---|---|\n| Pay for an HTTP resource that returns **402** (pay-per-call), search x402 bazaars, or monetize your own endpoint | `vendored\u002Fcoinbase\u002Fagentic-wallet\u002FSKILL.md` — Coinbase x402 client: read the price, pay, retry with proof; start here for a one-off paid request. `vendored\u002Fchaingpt\u002Fx402\u002FSKILL.md` — x402 discovery\u002Fmonetization: find paid services or put your own endpoint behind 402. |\n| Stream micropayments or run metered sessions (per-second \u002F per-token, not per-request) | `vendored\u002Ftempo\u002Fmppx\u002FSKILL.md` — Tempo\u002FStripe MPP charges, sessions, and streaming. Load when billing is continuous rather than a single call. **Not** for spending caps on per-request rails: a capped or recurring budget over x402 is `integrations\u002Fx402-erc7710\u002FSKILL.md`, not MPP. |\n| Dispatch a payment without hard-coding the rail (x402 vs MPP vs a2a-pay) | `vendored\u002Fokx\u002Fokx-agent-payments-protocol\u002FSKILL.md` — OKX OnchainOS unified payment dispatcher. Load when the counterparty's rail is unknown or may vary. |\n| Run a payment *through* a bounded permission (spend\u002Fsubscription policy on the rail) | `integrations\u002Fx402-erc7710\u002FSKILL.md` — connector fusing the x402 rail with an ERC-7710 delegated budget. Load for Guarded\u002FAdjudicated paid access. |\n\n### Authority, custody & delegation\n\n| When you need to… | Load |\n|---|---|\n| Grant an agent bounded, revocable authority (ERC-7710 delegations, ERC-7715 permission requests, caveats) | `vendored\u002Fmetamask\u002Fsmart-accounts-kit\u002Fskill.md` — MetaMask smart accounts + delegation mechanics. The Guarded-level instrument for on-chain authority. Also the row for the user **manually revoking** their own delegation. |\n| Give an agent a wallet with a built-in policy gate (per-tx caps, velocity limits, session keys) instead of a raw key | `vendored\u002Fchaingpt\u002Fagent-wallet\u002FSKILL.md` — custody-free policy-gated wallet. Load when the agent must sign but you won't hand over a private key. |\n| Provision or authenticate embedded\u002Fserver wallets, with policy-gated signing across chains | `vendored\u002Fprivy\u002Fprivy\u002FSKILL.md` — Privy wallet identity + auth. |\n\nThe first two are **mutually exclusive — pick by who holds the key**: the user\nkeeps custody and delegates → `smart-accounts-kit`; the agent gets its own\nwallet (guardrails built in) → `agent-wallet`. Never load both for one design.\n\n### Identity, reputation & negotiation\n\n| When you need to… | Load |\n|---|---|\n| Register, discover, or vet an agent on-chain (ERC-8004 identity, reputation, validation) | `vendored\u002Fchaingpt\u002Ftrustless-agents\u002FSKILL.md` — ERC-8004 registries; the default identity\u002Freputation skill. Also mintable via `vendored\u002Fopenserv\u002Fopenserv-multi-agent-workflows\u002FSKILL.md`, on Starknet via `vendored\u002Fstarknet\u002Fstarknet-identity\u002FSKILL.md`, on BNB via `vendored\u002Fbnb-chain\u002Fbnbchain-mcp\u002FSKILL.md`. For a human-readable named-account identity anchor (`alice.near`), `vendored\u002Fnear\u002Fnear-api-js\u002FSKILL.md`. |\n| Prove a human stands behind the agent | `vendored\u002Fhumanode\u002Fhumanode-agentlink\u002FSKILL.md` — Humanode AgentLink: sign HTTP requests, on-chain registry. |\n| Talk to another agent: agent cards, offers, task lifecycle (A2A) | `vendored\u002Fterminalskills\u002Fa2a-protocol\u002FSKILL.md` — A2A protocol; load at the negotiation step, before terms lock. For orchestrating many agents: `vendored\u002Fopenserv\u002Fopenserv-multi-agent-workflows\u002FSKILL.md`. |\n\n### Contracts & escrow\n\n| When you need to… | Load |\n|---|---|\n| Lock funds that release on an arbiter's decision (conditional escrow) | `vendored\u002Farkhai\u002Falkahest-user\u002FSKILL.md` — Alkahest EAS-based escrow with arbiters; the default escrow instrument. To **build on** Alkahest (write custom arbiters\u002Fobligations, integrate it), `vendored\u002Farkhai\u002Falkahest-developer\u002FSKILL.md`. |\n| Escrow behind a plain-language demand an AI oracle judges | `vendored\u002Farkhai\u002Fnla-create\u002FSKILL.md` (draft the agreement), `vendored\u002Farkhai\u002Fnla-fulfill\u002FSKILL.md` (deliver against it), `vendored\u002Farkhai\u002Fnla-arbitrate\u002FSKILL.md` (render the verdict) — natural-language-agreement escrows. |\n| Bounty escrow that pays out for making a failing test suite pass | `vendored\u002Farkhai\u002Fmake-git-escrow\u002FSKILL.md` (post the bounty), `vendored\u002Farkhai\u002Ffulfill-git-escrow\u002FSKILL.md` (claim it) — git-escrow, for code-delivery deals settled by tests. |\n\n### Verification, adjudication & disputes\n\n| When you need to… | Load |\n|---|---|\n| Write \u002F deploy \u002F test \u002F lint a GenLayer Intelligent Contract | `vendored\u002Fgenlayer\u002Fwrite-contract\u002FSKILL.md` (author), `vendored\u002Fgenlayer\u002Fgenlayer-cli\u002FSKILL.md` (deploy & call), `vendored\u002Fgenlayer\u002Fdirect-tests\u002FSKILL.md` + `vendored\u002Fgenlayer\u002Fintegration-tests\u002FSKILL.md` (test), `vendored\u002Fgenlayer\u002Fgenvm-lint\u002FSKILL.md` (lint). Load the one matching your build step. |\n| Design the adjudication itself: review rubric, evidence schema, decision payload | `integrations\u002Fgenlayer-intelligent-contracts\u002FSKILL.md` — the Adjudicated-level review interface. Load when the outcome needs qualitative \u002F natural-language judgment. |\n| Turn a verdict into enforcement (revoke or constrain an ERC-7710 permission) | `integrations\u002Fgenlayer-erc7710-connector\u002FSKILL.md` — relayer\u002Fcontroller wiring a decision to on-chain revocation. **Verdict-driven only**: load when adjudication must change the agent's authority; a user manually revoking their own delegation needs only `smart-accounts-kit`. |\n| Settle a narrow binary question from public web evidence (prediction market \u002F factual oracle) | `vendored\u002Fintelligent-oracle\u002Fintelligent-oracle\u002FSKILL.md` — Intelligent Oracle. Refetch `https:\u002F\u002Fwww.intelligentoracle.com\u002Fskill.md` before schema-sensitive work. |\n| Third-party arbitration as an alternative to GenLayer (**disclose** it isn't GenLayer's own) | `vendored\u002Fkleros\u002Fkleros-curate\u002FSKILL.md` (token-curated registries \u002F challenges) + `vendored\u002Fkleros\u002Fkleros-ipfs-upload\u002FSKILL.md` (IPFS evidence); or `vendored\u002Farkhai\u002Fnla-arbitrate\u002FSKILL.md` for LLM\u002Fmanual NLA arbitration. |\n\n### Execution, data & value movement (layer 5)\n\n| When you need to… | Load |\n|---|---|\n| Buy or sell AI inference (compute execution) | Pick **one** by venue: `vendored\u002F0g\u002F0g-compute\u002FSKILL.md` (decentralized compute\u002Fstorage\u002Fverifiable inference — the default when the venue is open), `vendored\u002Fantseed\u002Fantseed-connect\u002FSKILL.md` (P2P inference over USDC channels), `vendored\u002Fheurist\u002Fheurist-mesh-skill\u002FSKILL.md` (Heurist Mesh agents, x402 pay-per-call), `vendored\u002Fnear\u002Fnear-ai-cloud\u002FSKILL.md` (private inference with **TEE attestation** — verifiable-execution evidence). If the venue is not yet known, load only the 0G default and ask — do not preload the rest. |\n| Move value across chains (swaps \u002F bridging) | `vendored\u002Flifi\u002Flifi\u002FSKILL.md` — LI.FI cross-chain routing; `vendored\u002Flifi\u002Flifi-stablecoin-swap\u002FSKILL.md` for stablecoin-specific swaps; `vendored\u002Fnear\u002Fnear-intents\u002FSKILL.md` — NEAR Intents: cross-chain fulfilment via solver competition (chain-abstracted, one agent acts across chains). |\n| Pull on-chain data as evidence (balances, tx history, labels, holders, smart-money) | `vendored\u002Fchainbase\u002Fweb3-data\u002FSKILL.md` — general on-chain data. For deeper intelligence, the **Nansen** 7-skill pack under `vendored\u002Fnansen\u002F`: start with `nansen-token-research`, then `nansen-wallet-profiler` \u002F `nansen-holder-analysis` \u002F `nansen-smart-money-tracker` \u002F `nansen-general-search` \u002F `nansen-prediction-markets` \u002F `nansen-mpp-payment` as the question narrows. |\n\n### Full-platform packs (catch-all entry, then narrow)\n\n| When you need to… | Load |\n|---|---|\n| Anything on **OKX OnchainOS** (wallet, DEX data, DeFi, payments, agent-commerce) | 8-skill pack under `vendored\u002Fokx\u002F`. Start with `okx-ai` — the agent-commerce entry (ERC-8004 identity + task marketplace **with disputes** + A2A chat), the most Internet-Court-relevant. Then narrow: `okx-agentic-wallet` (swap\u002Fbridge\u002Fstrategy\u002Fsecurity), `okx-dex` (read-only DEX data), `okx-defi` (invest\u002Fportfolio), `okx-agent-payments-protocol` (x402\u002FMPP\u002Fa2a-pay), `okx-dapp-discovery` (route to third-party dapps), `okx-guide` (onboarding\u002Fsupport). |\n| **ChainGPT** AI tooling (contract gen\u002Faudit, news) + 140-tool MCP | `vendored\u002Fchaingpt\u002Fchaingpt\u002FSKILL.md` is the catch-all hub. Its focused siblings are routed above: `x402`, `agent-wallet`, `trustless-agents`. |\n| **AltLayer** AltLLM portal + Cloud-Claw agent VMs | 7-skill pack under `vendored\u002Faltlayer\u002F`. Entry: `altllm-portal-cli` (portal ops) and `cloud-claw` \u002F `cloud-claw-launch-agent` (spin up agent VMs). The rest are plumbing: `altllm-portal-auth`, `-api-keys`, `-billing`, `-payments`. |\n| **OpenServ** multi-agent orchestration + SDK (mints ERC-8004 identities) | 5-skill pack under `vendored\u002Fopenserv\u002F`. Entry: `openserv-multi-agent-workflows`. Others: `openserv-agent-sdk`, `openserv-client`, `openserv-launch`, `openserv-ideaboard-api`. |\n| **Starknet** L2 (Cairo, native account abstraction) | 4-skill pack under `vendored\u002Fstarknet\u002F`. Entry: `starknet-js` (SDK). Others: `starknet-identity` (ERC-8004), `starknet-defi`, `starknet-wallet`. |\n| **BNB Chain** ops incl. ERC-8004 registration & Greenfield storage | `vendored\u002Fbnb-chain\u002Fbnbchain-mcp\u002FSKILL.md`. |\n| **NEAR** — AI-native L1 (Shade Agents \u002F TEE inference, named-account identity, chain-abstracted payment) | 6-skill pack under `vendored\u002Fnear\u002F`. Reach for the on-theme ones first: `near-intents` (cross-chain payment\u002Ffulfilment), `near-ai-cloud` (TEE-verifiable inference as evidence), `near-api-js` (chain interaction + named-account identity). Also: `near-kit` (type-safe TS SDK + sandbox testing), `near-smart-contracts` (Rust contracts), `near-dapp` (full-stack React\u002FNext.js + wallet). |\n\n## Workflow\n\n1. Classify the deal:\n   - Delegated agent authority or revocable mandate.\n   - Pay-per-use API or content access.\n   - Recurring spend or subscription.\n   - Escrowed service delivery.\n   - Web-evidence oracle or prediction market.\n   - Disputed job, SLA, or refund.\n2. Extract commercial terms:\n   - Parties, objective, authority, prohibited actions, amount, cadence, max\n     spend, expiration, deliverable, evidence source, review cadence,\n     revocation path, refund path, and dispute window.\n3. Select rails (see the stack table):\n   - x402 for immediate HTTP payment.\n   - ERC-7710 plus ERC-7715 when an agent needs bounded delegated capability.\n   - MPP sessions for streamed micropayments; AP2 mandates for\n     user-authorized purchases.\n   - Check for a **chain mismatch** between where the funds sit and where the\n     payee is paid — if they differ, the plan also needs the value-movement\n     row (LI.FI \u002F NEAR Intents) before any payment step.\n   - Escrow (ERC-8183, Arkhai) when funds should stay locked until a\n     fulfillment attestation satisfies an arbiter.\n   - GenLayer Intelligent Contracts when the outcome depends on qualitative\n     review of agent behavior or natural-language judgment.\n   - Intelligent Oracle when the outcome is a narrow binary question settled\n     from public web evidence.\n   - GenLayer Intelligent Contracts also when a disputed job, SLA, or\n     refund needs an independent verdict.\n4. Produce a flow artifact:\n   - User story, sequence steps, data model, permission policy, review\n     contract, evidence schema, revocation path, failure cases, and minimal\n     implementation plan.\n5. Add dispute handling:\n   - Define what can go wrong, who can trigger a dispute, what evidence is\n     accepted, how finality is reached, and what happens to funds.\n\n## Revocable Agent Mandates\n\nUse this pattern when the user delegates authority for a meaningful period and\nwants to constrain or revoke it if the agent underperforms. The outputs, from\nplain language down to enforcement:\n\n1. Natural-language mandate — what the agent may and may not do.\n2. Machine-readable policy object capturing the same limits.\n3. A bounded on-chain permission that grants only that authority (for example,\n   an ERC-7710 delegation with caveats).\n4. A review rubric an independent adjudicator applies to the agent's actions.\n5. An action-receipt and evidence schema the review consumes.\n6. A path from verdict to enforcement — how a review decision constrains or\n   revokes the granted authority (for example, a controller wired to the\n   permission).\n7. Failure and appeal paths.\n\nMatch each output to the trust level and rails chosen for the deal; not every\nmandate needs all seven. Do not claim a reviewer can directly cancel a\npermission unless a revocation controller is actually part of the design. For\nlong-lived authority, prefer an active revocation controller plus an absolute\nexpiry as a fail-safe.\n\n## Wallet-UI Deployment Mode\n\nWhen the user deploys contracts, grants permissions, or relays decisions from\nMetaMask, WalletConnect, or another user-controlled wallet UI:\n\n- Never ask for a private key, seed phrase, or unrestricted bearer wallet.\n- Never claim a contract, wallet, delegation, permission, review, relay, or\n  payment exists until a proof arrives: transaction hash (`0x` + 64 hex),\n  address (`0x` + 40 hex), signed artifact, or receipt. Malformed proof →\n  ask again; do not advance.\n- If the user says a proof is simulated, you may continue a rehearsal, but\n  label that state as simulated in every later mention.\n- Build transaction cards from source, ABI, or deployment metadata — never\n  invent constructor fields from contract names. A card includes chain,\n  sender, target, value, exact arguments, expected post-transaction reads,\n  and the proof needed before proceeding.\n- Do not paste full bytecode or calldata into chat; show the selector,\n  payload hash, byte length, and exact arguments instead.\n- End your turn while waiting for a user approval or signature — the approval\n  is the user's act in their wallet UI, and the chat is only the command\n  surface. Never promise background polling, and never advance a step until\n  its proof arrives.\n\n## Output Shape\n\nWhen designing a flow or integration, return:\n\n1. Deal summary in plain English.\n2. Stack selection and why each rail is used.\n3. Sequence diagram in text form.\n4. Data model or policy object.\n5. Happy path.\n6. Failure and dispute paths.\n7. Smallest credible implementation plan.\n8. Open assumptions and test plan.\n\n## Guardrails\n\n- Treat Internet Court as protocol and integration design, not legal advice.\n- Do not assume a payment rail provides escrow, refunds, or chargebacks. Pair\n  it with escrow or adjudication when performance risk matters.\n- Treat draft or unproven standards as draft — confirm the target stack\n  actually supports a permission or delegation scheme before relying on it.\n  Simulate execution before doing it for real; design for revocation,\n  expiration, and stale permissions.\n- Never recommend unlimited token approvals or unbounded agent spend.\n- Keep integrations testnet-first unless the user explicitly requests\n  production.\n- Report failures faithfully — a refused payment after revocation is the\n  system working, not an error to hide.\n\n## Reporting Coverage Gaps\n\nInternet Court is an evolving package, and its coverage is not complete. When\nyou hit a wall — a legitimate agent-commerce use case you cannot complete\nbecause no skill in this package handles it, a routed skill is missing, broken,\nor out of date, or a capability works poorly — do **both** of the following:\n\n1. Tell the user plainly what is missing or failing. Never fabricate a result,\n   never silently route around the gap, and never claim something works when it\n   does not.\n2. Open an issue on the package repo so maintainers can add or fix coverage:\n   **`https:\u002F\u002Fgithub.com\u002Finternet-court\u002Finternet-court-skill\u002Fissues`**. If you\n   have a GitHub tool available (for example `gh issue create`), file it\n   directly; otherwise hand the user a ready-to-file title and body and the\n   link above. Search existing issues first and don't file duplicates — one\n   issue per distinct gap.\n\nPut this in the issue:\n\n- the use case \u002F user goal, in one line;\n- which layer or need it maps to (reference the stack and routing tables);\n- what is missing or failing — no skill for X, skill `Y` errored, or `Z`\n  produced a wrong or low-quality result — with the exact skill path when\n  there is one;\n- what a fix would look like: a skill to vendor, a connector to add, or a doc\n  to update.\n\nFiling the gap is part of doing the job well: it is how the package learns\nwhat to cover next.\n",{"name":5,"description":7},"master","63f17866001c5ca854bf2fa0b678ce277133de5b","skills\u002Fmarketplace\u002Finternet-court",0,"https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS","https:\u002F\u002Fbg-szy.github.io\u002FTOP-SKILLS\u002F","https:\u002F\u002Fgithub.com\u002Fbg-szy","https:\u002F\u002Fwww.ecust.edu.cn\u002F","a student of East China University of Science and Technology","2026-11-15 03:38:02",{"source":8,"skillName":5,"commands":36,"prompt":46,"zipAvailable":47},[37,40,43],{"tool":38,"command":39},"npx","npx skills add bg-szy\u002FTOP-SKILLS --skill internet-court",{"tool":41,"command":42},"bunx","bunx skills add bg-szy\u002FTOP-SKILLS --skill internet-court",{"tool":44,"command":45},"pnpm","pnpm dlx skills add bg-szy\u002FTOP-SKILLS --skill internet-court","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Finternet-court\u002FSKILL.md",true,[49,53,59,63],{"path":50,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":52,"sourceHtmlUrl":16},"LICENSE","许可证",1082,{"path":54,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":57,"sourceHtmlUrl":58},"README.md","REFERENCE","参考文档",13035,"https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Finternet-court\u002FREADME.md",{"path":60,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":61,"sourceHtmlUrl":62},"skill-report.json",520698,"https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Finternet-court\u002Fskill-report.json",{"path":64,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":65,"sourceHtmlUrl":66},"skills-lock.json",32798,"https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Finternet-court\u002Fskills-lock.json",[68,76,83,90,97,104],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":75},"01M7DXW9N826PKXK174PVW3038","serpmaven","Scrape Google search results (SERP) to JSON with SERP Maven (serpmaven) - a real Chrome browser on a residential proxy with sticky sessions. Run it as a CLI, a Python library, a local HTTP search API (`serve`), an MCP server or a Docker image; set the proxy login, workers, pages and cost tracking. Use when the user wants Google results, rank tracking, a self-hosted SERP API, or live Google search for an AI agent.","https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Fserpmaven\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":75},"01M7DXVHSTWDV65475QEQ915M7","revnu-landing-page-audit","Find out why a landing page isn't turning visitors into signups, demos or sales. Audits one page against what the ad, post or search that sends people there promised, then checks the first screen, proof, call to action, friction and mobile. Returns at most 7 prioritized fixes, plus what is fine and what could not be checked; rewrites specific lines and fixes the page in the repo when asked. Never promises a conversion lift. Works without a Revnu account. Use when the user says or asks: \"audit my landing page\", \"why isn't this page converting\", \"review my homepage\", \"my ads get clicks but no signups\", \"people bounce from the pricing page\", \"CRO review\", \"improve conversion rate\", \"is my hero section clear\", \"roast my landing page\", \"rewrite my headline\", \"what's wrong with this page\", \"landing page feedback\", \"make the signup page convert better\". Not for SEO or AI-search audits, page speed engineering on its own, or building a new page from scratch.","https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Frevnu-landing-page-audit\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":75},"01M7DXV8X2D7B3VTD4HHYYHTGR","revnu-growth-context","Write down once who buys, what they're stuck on, what it costs, what proof you have and what you've already tried, so every growth task starts from facts instead of a questionnaire. Creates or updates .agents\u002Frevnu-growth.md from the repo, the live site and the user, and imports an existing product-marketing.md. Use when the user says \"set up our marketing context\", \"who is our ideal customer\", \"write our ICP\", \"describe the product for marketing\", \"positioning doc\", \"what's our brand voice\", \"remember this about our customers\", \"log what we tried\", \"what have we learned from marketing\", \"we changed pricing, update the context\", \"update our growth notes\", \"summarize the business for the agent\", \"import my product-marketing.md\", \"start a growth doc\", or before a first marketing, SEO, outreach or launch task when no context file exists. Works without a Revnu account.","https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Frevnu-growth-context\u002FSKILL.md",[],[],{"skillNo":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"publishedAt":75},"01M7DXV0DGJ1G0A60PAHC6J3C5","revnu-get-listed","Get onto the third-party pages AI answers and search cite when buyers ask for the best tool in your category: \"best X\" roundups, directories, review sites, GitHub awesome lists, community threads and comparison pages. Finds which ones are cited, ranks them by fit and effort, and drafts each submission, PR or pitch by the target's own rules. Use when the user says \"get us listed\", \"which directories should we submit to\", \"how do we get into best-of lists\", \"why does ChatGPT recommend competitors and not us\", \"get mentioned in AI answers\", \"submit to awesome lists\", \"get on G2\", \"where are our competitors listed\", \"pitch listicle authors\", \"find backlink opportunities\", \"launch directories\", \"get into roundups\", \"Product Hunt alternatives\", \"where should we be listed\", \"get cited by Perplexity\". Never submits, pays or posts without a yes. Works without a Revnu account.","https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Frevnu-get-listed\u002FSKILL.md",[],[],{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":103,"publishedAt":75},"01M7DXTSPMGT9D6DSS0HP9BSTA","revnu-ai-readiness-audit","Audit whether your website gives ChatGPT, Perplexity, Claude, Gemini and Google's AI Overviews something worth quoting, and fix the pages that don't. Use when the user says or asks \"is my site ready for AI search\", \"audit my site for AI answers\", \"AEO audit\", \"GEO audit\", \"why doesn't ChatGPT quote our pricing\", \"make our pages citable by AI\", \"do I need llms.txt\", \"does schema help with ChatGPT\", \"check our structured data\", \"is our content visible without JavaScript\", \"AI SEO audit\", \"optimize our homepage for AI answers\", \"LLM SEO check\", \"why does AI get our product wrong\", \"fix our site for AI Overviews\". Checks fetchability and indexability, answers in the raw HTML, entity facts, schema (including JS-injected markup), headings, dates and llms.txt, then gives at most 10 prioritised fixes and edits the repo when asked. Works without a Revnu account.","https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Frevnu-ai-readiness-audit\u002FSKILL.md",[],[],{"skillNo":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":108,"categories":109,"keywords":110,"publishedAt":75},"01M7DXSZHCVTRS620GC3S97JZV","proxy-troubleshooting","Diagnose residential proxy failures, with NodeMaven's gateway as the reference - 407 with correct credentials, 406, 410, 500, a connection that hangs, ERR_TUNNEL_CONNECTION_FAILED in a browser, Playwright or Puppeteer ignoring proxy credentials, a sticky session that changes IP, geo-targeting that is silently ignored, passwords with special characters, and unexpected traffic from headless Chrome. Use when a proxy connection fails, returns the wrong location, or costs more traffic than expected.","https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Fproxy-troubleshooting\u002FSKILL.md",[],[]]