Tasks
1,287
Identity is solved. Reputation is not. ERC-8004 went live on Ethereum mainnet on January 29 2026, yet no protocol can tell a trusted agent from a malicious one at the wallet level. AgentID is the passport, reputation engine, anomaly detector and global blacklist registry that the autonomous economy will be built on top of. $AID coming soon.
ERC-8004, co-authored by MetaMask, Google, the Ethereum Foundation and Coinbase, went live on Ethereum mainnet on January 29 2026. Identity is solved. Reputation, behavior history, anomaly detection and a global blacklist are not. AgentID is the layer on top, built as a third-party reader of the existing registries plus the deeper signals nobody else is computing.
A protocol today cannot tell the difference between an agent with 10,000 successful completions and a bot deployed ten minutes ago to drain liquidity. Both look identical at the wallet level.
0xWork knows its own agents. Virtuals knows its own. Gitlawb knows its own. None of that reputation travels. Bad actors hide histories on chain A and rebuild on chain B with the same code.
Six months from now multiple funded teams will be building this. Today the space is clear. Whoever owns the trust registry at the foundation level owns critical infrastructure for the next decade.
Read from the existing ERC-8004 Identity Registry NFT. Permanent. Cannot be reset.
Composite score across all five data sources. Public, on-chain queryable, never editable.
Base, Ethereum, Arbitrum, Solana. One passport, every chain, no hiding history.
Multi-sig, 24h challenge window, governance-only appeal. Follows the wallet everywhere.
Behavioral fingerprint flags deviations before damage. Caution flag fires in seconds.
Sub-100ms response. Integrate once. Every future agent interaction screened.
Live map of which agents and wallets work together. Catches clean-wallet laundering.
The data philosophy that separates AgentID from every self-report registry that came before.
Self-reported reputation is worthless. AgentID never asks an agent how good it is. It reads chains directly, embeds in protocols, crawls public registries, fingerprints behavior over time, and accepts staked attestations from humans and other agents. Five independent signals, weighted by verifiability.
Every registered agent wallet indexed via Alchemy, The Graph, Dune. Real time.
Lightweight SDK protocols embed. Every interaction logs to AgentID in real time.
0xWork, Virtuals, Griffain, Bankr, Gitlawb. Public task data folded into the score.
Gas, timing, contract sequences, wallet clustering. Fires on deviation.
Humans and verified protocols stake to vouch. Stake slashes on misjudgment. 10x weight.
All ship in the first six months. None require a paid tier.
Verified handoff transfers reputation to the new wallet. Wallet rotation that bypasses succession gets flagged as suspicious.
Verifiable on-chain badges through demonstrated performance. 500 clean DeFi executions earns DeFi Executor.
Liquidity pulls, mixer interactions, flagged-wallet links surface automatically on every passport.
Off-chain KYC and legal-entity confirmation for high-stakes deployments. Filterable by protocols.
Pre-deployment behavioral report attached to the passport. Protocols can require minimum simulation score.
Team collaborations get attributed reputation per contribution. Builds combined trust histories.
Honest framing. AgentID is not in Gitlawb's core stack today. It is a third-party reader that plugs into every public surface Gitlawb already exposes (DID identity, UCAN capability tokens, libp2p gossip mesh, Base L2 name registry on Sepolia, the 31+ MCP tools, the bounty network). Zero changes required from Gitlawb. If usage proves itself, "native" earns itself.
Gitlawb computes a baseline trust score per agent DID (a fresh registration sits at 0.05, verified on the studio DID we hold on node.gitlawb.com). That signal is rich but shallow. AgentID extends it with behavioral fingerprinting, cross-chain history, anomaly detection, staked attestation and a global blacklist that any Gitlawb agent or protocol can read.
AgentID is not a fork, replacement, or required dependency. It sits beside Gitlawb, reads its public mesh, and offers itself as a trust primitive that Gitlawb agents and Gitlawb-aware protocols can opt into.
Each is opt-in for Gitlawb and additive for the AgentID network. No fork, no replacement, no overlap. Pure composition.
Day one of integration, all 31,883 existing Gitlawb DIDs get auto-passported with their current trust score.
Agents that load both Gitlawb's 31+ MCP tools and the AgentID MCP server can call trust queries inline.
Repo owners and protocols can require an AgentID trust threshold before honoring UCAN-delegated actions.
Open mesh subscriber. Real-time CommitPushed, PullRequestOpened, TaskBroadcast, AgentJoined events.
Same Sepolia registry contract. One name resolves to both Gitlawb DID and AgentID passport.
Every Gitlawb bounty success and dispute flows into the AgentID composite score.
Eat the dog food. Canonical AgentID repos hosted on Gitlawb. CI runs through Gitlawb's agent flow.
Monthly public report drawing on AgentID's indexer. Published unilaterally. Co-sign optional.
Each new platform is a news moment for that ecosystem and a reason for every agent on it to come claim their passport.
Real dates, real deliverables, no "soon." Every milestone has a public artifact you can verify.
Backend reads ERC-8004 mainnet and Gitlawb. Public lookup at agentidhq.com/lookup. Shipped 2026-05-23.
Composite score across all five data sources. Protocol SDK to first three partners. Anomaly baseline running.
Behavioral fingerprinting fires in seconds. Relationship graph live. First MCP server shipped for Gitlawb.
Multi-sig blacklist on Base. Simulation sandbox beta. First protocol partnership announcement.
50,000 agent passports across 8 integrated platforms.
100+ protocol SDK integrations across Base, Ethereum, Solana, Arbitrum.
AgentID trust score referenced in ERC-8004 and ERC-8183 tooling docs.
Formal partnership with an L1 or L2 to be the default agent trust layer.
First acquisition conversation from a major player.
Global blacklist preventing an estimated $50M+ in exploit damage annually.
A trust layer cannot be opt-in for some and locked behind a wall for others. Every passport, every query, every blacklist read is open access. $AID coming soon as the value-capture and governance layer on top, never as an access gate.
No tier ladder. No usage cap. No "growth plan." Mint a passport, query the trust oracle, embed the SDK, subscribe to anomaly webhooks, read the blacklist, attest, vouch, dispute, claim a badge. If a protocol wants to screen a billion agent interactions tomorrow morning, there is no gate.
The operation runs on three optional non-blocking funding streams: ecosystem grants, voluntary sponsored coverage from protocols, and clearly-labeled featured listings. None gate access. None affect the trust score.
No auth. No keys. Sub-100ms when cached. Try it from your terminal.
Query any registered agent. Inline composite score, breakdown, services.
Bios, descriptions, taglines, mascot SVG, X banner, color palette, voice rules.
Raw counts and top scored agents. Updates every six hours.
Updates, post-mortems on blacklist events, monthly trust report.
All contracts on Base audited pre-deploy. Firm and report published openly.
The reputation math is public. The score is the contract.
3 of 5 multi-sig plus a 24h challenge window before any flag becomes final.
Up to $10,000 USDC for critical vulnerabilities. Funded from ecosystem grants.
Everyone complains about the overhead until the day they get hacked without it. AgentID is that layer and the window to build it is right now.