Archon Competitive Analysis

Last updated: 2026-10-15 13:30 EDT
Refresh cycle: Weekly during evangelism sweeps, ad-hoc for new discoveries
Maintained by: Morningstar

Overview

This research project tracks decentralized identity initiatives for AI agents, monitoring the competitive landscape around Archon. The goal is to understand market positioning, identify differentiators, and surface collaboration or integration opportunities.

Methodology

Discovery sources

  • GitHub search: “agent identity DID”, “verifiable credentials AI agent”, “decentralized identity AI agents”, “agent identity protocol”
  • GitHub REST API repo metadata checks
  • README/manual repo inspection for current project framing
  • Standards bodies: W3C DID & VC work, Internet-Drafts
  • Community chatter around agent identity, messaging, payments, and trust infrastructure

Evaluation criteria

  • Standards compliance (W3C DID Core, VC Data Model, Bitstring Status List)
  • DID method or cryptographic identity primitive
  • Scope (identity-only vs. authorization vs. messaging vs. compliance vs. commerce)
  • Architecture (centralized service, relay, blockchain, web-federated, decentralized P2P)
  • Maturity (stars, recent pushes, docs quality, repo availability)
  • Agent-specific features (delegation, receipts, persistence, substrate independence, trust signals)

Update cadence: Weekly evangelism sweeps, plus ad-hoc refreshes when new projects surface.

Projects Tracked

Project Stars Language Identity / Auth Primitive Scope Status
Bindu 10058 Python platform-administered did:bindu + mTLS + Hydra OAuth + x402 Identity, A2A communication, inbox, payments, gateway 🚨 Highest-traction direct/adjacent stack + bridge opportunity; first star decline 10076→10058 (-18); no repo push since 2026-09-06 (sixth week)
Urbit 3622 / 80 Hoon / C Urbit ID / Azimuth PKI + Ames networking Personal server OS, P2P network, decentralized identity ✅ High-traction protocol/substrate incumbent
Agent Network Protocol (ANP) 1450 HTML/docs did:wba Open agent communication protocol suite ✅ Protocol/spec leader
AgentConnect 352 Rust did:wba authentication ANP SDK / implementation ✅ Implementation path to watch; primary language shifted Python→Rust
Agent-Safe Pipeline 688 TypeScript independent policy-verdict boundary + single-use intent-bound grants Authorization boundary / execution gating ✅ 687→688★ (+1) — last cycle’s +98 spike flattened; pushed 2026-10-13
AgenticMail 237 TypeScript N/A Email/SMS/phone-call infra ✅ Strong adjacent traction
MolTrust N/A (commercial service) TypeScript/Node did:moltrust + did:web + ERC-8004, Base L2 anchor Agent trust infrastructure: identity, VC, reputation, mandates, audit 🆕 Most mature centralized commercial rival
Agent Passport System 49 TypeScript did:aps + accepts did:key/did:web/SPIFFE/OAuth Delegation narrowing, gateway enforcement, signed receipts ✅ Flat at 49★, pushed 2026-10-15; IETF draft -04 (2026-09-28)
decern 13 Rust AuthZEN PDP + SMT-proven invariants + signed hash-chained ledger Deterministic authorization + tamper-evident audit ✅ Flat at 13★, pushed 2026-10-14
SIQ Agent Security 50 Python Ed25519-signed intent/parameter/receipt binding + Skill Execution Context Agent/skill runtime authorization boundary + enterprise control plane ✅ 51→50★ (-1), pushed 2026-10-09
Grantex 33 TypeScript delegated authorization / commerce passport Agent authorization + audit + commerce ✅ High-signal watchlist item; flat at 33★, pushed 2026-10-15
Attestix 874 Python did:key / did:web Compliance + credentials + MCP ⚠️ Flat at 874★; pushed 2026-10-15 (resumed commits after two-week pause) — traction still unverified
AgentNexus 10 Python DID + relay + encrypted messaging Agent team communication, workflow, artifacts, authorization ✅ Collaboration substrate watchlist
Kestrel Sovereign 8 Python portable DID identity Sovereign agent framework + memory + governance ✅ Sovereign-agent framework pressure
Agentic Airlock 2 Python Ed25519 + OAuth 2.1 + trust score claims Trust/compliance layer, delegation chains, audit ✅ OAuth/compliance-oriented watchlist
Chancery 26 Go registry + scoped delegation + in-path MCP enforcement Agent IdP, instant revocation, tamper-evident audit ❌ Inactive: flat at 26★, no push since 2026-07-21 (tenth stalled cycle)
AgentValet 2 TypeScript IETF AIMS + SPIFFE + AuthZEN + CIBA Identity governance + credential governance + MCP proxy ✅ Enterprise IGA / MCP proxy pressure (last push 2026-09-26)
Agent Name Service (ANS) 43 / 33 Go IETF-draft naming registry + Merkle transparency log + SCITT COSE receipts Agent naming/discovery + trust index 🆕 Registry/transparency-log pressure nearest Archon’s registry layer
Hedera / did:hedera 36 / 27 / 69 Java/spec/TypeScript did:hedera / HCS / HBAR / HTS DID method + VC SDK + agent/payment/audit substrate ✅ Direct DID competitor + high-signal adjacent substrate
Soulverse N/A Private / website claims did:soul currently proprietary / future decentralization under consideration + W3C DID/VC claims Pre-execution validation, agent governance, credential-gated execution 🆕 Agent-governance / execution-gating watchlist
didit skills 28 Python N/A KYC / verification API wrappers ✅ Adjacent, non-competitor
AIP 14 Python did:aip / Ed25519 Identity + trust chains + encrypted messaging ✅ Partial overlap
clawdentity 9 TypeScript did:cdi Cross-platform messaging + identity ✅ Closest philosophical rival
Motebit 7 TypeScript Ed25519 + signed receipts Sovereign agent runtime + trust routing ✅ Early but philosophically relevant
Credat 2 TypeScript scoped credentials Identity + delegation + verification SDK ✅ Small but directly relevant
HelixID 8 TypeScript DID + VC + scoped permissions Identity + authorization layer ✅ Standards-aligned watchlist; repo moved to helixid org
IDProva 1 Rust Ed25519 + delegated authority + receipts Enterprise identity/audit layer ✅ Watchlist
A2AL 1 Go cryptographic AID Decentralized agent discovery/networking ✅ P2P watchlist
Chorus 1 Rust decentralized identity + libp2p P2P encrypted communication layer ✅ P2P watchlist
payelink-agent-identity-sdk 2 Python did:key DID SDK only ✅ Narrow identity component
agent-did 0 TypeScript did:key DID + VC toolkit ✅ Direct standards competitor, low traction
agent-identity-hub N/A N/A did:ethr (historical) Platform/orchestration layer ❓ Repo unavailable / 404

Project Profiles

Bindu

Repository: https://github.com/GetBindu/Bindu Stars: 10058 | Language: Python | DID Method: platform-administered did:bindu Last pushed: 2026-09-06 | Last checked: 2026-10-15 (first star decline: 10076→10058, -18, forks 669→670 — sixth consecutive week with no repo push) Companion template: https://github.com/GetBindu/create-bindu-agent — 32★, Python, last pushed 2026-03-13

Bindu is now the highest-traction project in this landscape by a wide margin. GitHub describes it as “the identity, communication, and payments layer for AI agents.” The README frames the product as one-call plumbing: wrap an agent handler with bindufy() and it comes online with cryptographic identity, A2A JSON-RPC, optional public tunneling, and x402 USDC payment gating.

The strategic point is not just the star count. Bindu has the most complete DX wedge among the tracked projects: Python package, TypeScript/Kotlin SDK story, cookiecutter template, inbox/operator UI, gateway, skills, negotiation endpoint, storage/scheduler docs, payment middleware, and a polished docs site. It packages identity as part of a working agent-server stack instead of asking developers to care about DID theory first.

Evidence checked

  • GitHub API: GetBindu/Bindu has 7488★, 417 forks, 147 open issues/PRs, Python as primary language, default branch main, repo created 2025-03-16, pushed 2026-07-06, and homepage https://docs.getbindu.com
  • GitHub API: GetBindu/create-bindu-agent has 31★, 6 forks, MIT license metadata, and describes a cookiecutter template for Bindu agents
  • PyPI: package bindu current version 2026.21.1, requires Python >=3.12, summary “A protocol framework for agent-to-agent communication”
  • README/docs: Bindu uses A2A JSON-RPC, Ed25519 DID signatures, mTLS via Smallstep step-ca, OAuth2 via Ory Hydra, x402 payments, skills, private skills, negotiation, inbox UI, and gateway orchestration

Architecture / trust model

  • Identity: custom did:bindu:<author>:<name>:<uuid> method backed by Ed25519 keys
  • DID document storage/resolution: Bindu docs say the DID document lives inside the Hydra OAuth client’s metadata field, with a public resolver endpoint for A2A resolution
  • Authorization: agents self-register in Hydra; the DID is the OAuth client ID; bearer tokens are introspected through Hydra and last about one hour
  • Transport: mTLS with X.509 certs from step-ca, SAN tied to the DID, 24-hour TTL, auto-renewed in process
  • Message integrity: JSON bodies carry Ed25519 DID signatures in headers such as X-DID-Signature / X-DID-Identity
  • Payments: x402 USDC payment middleware; README claims Base, Base Sepolia, Ethereum, Ethereum Sepolia, and SKALE Europa pre-configured, with additional EVM chains configurable
  • Operator UX: inbox presents A2A messages between DIDs like email threads and verifies signatures inline

Strategic read Bindu is a serious competitive signal because it makes the agent-commerce stack legible: agent identity, communication, authentication, payment gating, and operator inbox in one framework. That directly pressures Archon’s public narrative. If developers encounter Bindu first, they may assume did:bindu + Hydra + A2A + x402 is “good enough” identity infrastructure because it solves the immediate packaging problem.

The weakness is sovereignty. Bindu’s DID method appears bound to Bindu/Hydra client metadata and Bindu-operated auth/resolver assumptions. That is useful operationally, but it should not be treated as a decentralized root of authority equivalent to did:cid. Bindu’s architecture is closer to a strong application-layer agent platform with cryptographic controls than a substrate-independent decentralized identity registry.

Collaboration opportunity

Bindu should be approached as a high-signal collaboration target, not only as a competitor. The useful split is simple: Bindu has strong agent-server DX, A2A/x402 packaging, inbox/operator UX, and payment middleware; Archon has the stronger candidate for sovereign root authority via did:cid, credentials, decentralized registry/discovery, and Lightning-native settlement adjacency. A practical collaboration path would let a Bindu agent use did:cid as its root identity, advertise or resolve through Bindu-style A2A flows, sign requests with Archon authority, and settle paid actions through whichever payment rail fits the use case.

The bridge thesis: Bindu can remain the app/platform rail; Archon can provide portable root authority and verifiable delegation. That avoids a false zero-sum framing while keeping the sovereignty distinction clear.

Archon comparison

  • Direct overlap: agent DID, DID resolution, signed messages, service discovery, credentials/skills, agent-to-agent protocol surface, payment-aware execution
  • Bindu advantage: developer packaging, docs, star traction, inbox UI, A2A/x402 familiarity, broad framework support, and production-shaped middleware
  • Archon advantage: sovereign did:cid, content-addressed identity, decentralized registry/discovery, substrate independence, W3C VC/status-list orientation, and Lightning-native payment adjacency
  • Recommended stance: treat Bindu as the top immediate narrative/DX competitor and the clearest near-term bridge partner. Archon should not dismiss it as “centralized DID”; instead, show concretely why a Bindu-style app platform benefits from a portable decentralized root of authority such as did:cid

Urbit

Primary repos: https://github.com/urbit/urbit and https://github.com/urbit/vere Stars: 3622 / 80 | Language: Hoon / C | Identity: Urbit ID / Azimuth PKI Last pushed: 2026-10-15 / 2026-10-15 | Last checked: 2026-10-15

Urbit is not an agent-DID product, but it is too structurally relevant to omit. Its docs define Urbit as a personal server, a peer-to-peer network of those servers, and a decentralized identity standard called Urbit ID. urbit/urbit is the high-traction core repo, while urbit/vere is the runtime layer containing the Nock VM, I/O drivers, event log, and snapshotting system.

The competitive pressure is substrate-level: Urbit already bundles sovereign identity, personal compute, peer networking, routing/discovery hierarchy, and application distribution into one coherent world. That makes it an incumbent for developers who think agents should live on personal servers rather than on SaaS agent platforms. It is adjacent to Archon rather than a direct W3C DID/VC competitor, but it competes for the same mental territory around persistent identity, autonomous services, and decentralized coordination.

Evidence checked

  • Urbit docs: Urbit is described as “a simple personal server (Urbit OS), a peer-to-peer network of those servers, and a decentralized identity standard (Urbit ID).”
  • Urbit ID docs: Urbit ID is described as decentralized, secure, human-meaningful, and based on a PKI implemented with NFTs on Ethereum; servers cryptographically sign network messages with the ID.
  • Azimuth docs: Azimuth.eth stores data related to Azimuth points and ownership, acting as the ledger for Urbit ID.
  • Ames docs: Ames is the networking protocol/vane; packets use crypto from /sys/zuse.hoon, peer public keys come from Jael PKI, and routing/discovery is built into the network.
  • GitHub API: urbit/urbit has 3615★, 364 forks, Hoon as primary language, and was pushed 2026-07-04; urbit/vere has 79★, 52 forks, C as primary language, and was pushed 2026-07-03.

Archon comparison

  • Direct overlap: decentralized identity, cryptographic signing, P2P discovery/routing, persistent agent/server identity, sovereign infrastructure narrative
  • Urbit advantage: mature cultural/protocol surface, personal-server OS, integrated P2P network, application ecosystem, and long-running identity hierarchy
  • Archon advantage: DID/VC-native agent authority, content-addressed did:cid, credential issuance/verification, service endpoints, and payment-aware delegated action without requiring adoption of a whole OS/network
  • Recommended stance: treat Urbit as a protocol/substrate incumbent to bridge or coexist with, not as a DID-method clone. Archon can position itself as a lighter agent authority layer that could bind Urbit-hosted services, Lightning wallets, Dmail, and external agent workflows under portable did:cid credentials and receipts

Agent Network Protocol (ANP)

Repository: https://github.com/agent-network-protocol/AgentNetworkProtocol Stars: 1450 | Language: HTML/docs | DID Method: did:wba Last pushed: 2026-10-15 | Last checked: 2026-10-15

ANP remains the highest-visibility protocol project in this landscape. Its repo description positions it as an open-source protocol for agent communication with a vision of an open, secure collaboration network for billions of intelligent agents.

Key features / signals

  • Still the visibility leader by a large margin
  • The primary repo remains documentation/spec-heavy rather than the main implementation codebase
  • The project now has an obvious implementation companion in AgentConnect
  • did:wba continues to be the main custom-method center of gravity outside Archon’s did:cid

Archon comparison

  • ANP is a broad communication protocol ecosystem; Archon is a sovereign identity substrate
  • ANP wins mindshare and protocol legibility; Archon differentiates on decentralized registry design, content-addressed identity, credentials, and sovereignty
  • Best treated as both a competitive narrative and a possible integration surface

AgentConnect

Repository: https://github.com/agent-network-protocol/anp Stars: 352 | Language: Rust (shifted from Python as of 2026-10-15) | Identity: DID-WBA authentication Last pushed: 2026-10-14 | Last checked: 2026-10-15

AgentConnect is the open-source SDK implementation path for ANP. Its README explicitly says it implements Agent Network Protocol and includes a DID-WBA authentication guide. Current README notes emphasize HTTP Message Signatures, Ed25519 Multikey binding keys, stricter resolver behavior, and access-token migration behavior.

Key features / signals

  • Material traction separate from the ANP spec repo
  • More implementation-relevant than the ANP docs repo alone
  • DID-WBA auth appears to be maturing toward concrete SDK behavior

Archon comparison

  • Important because ANP is no longer just a spec narrative; it has a visible SDK path
  • Archon should explain whether it complements, bridges to, or competes with DID-WBA auth in agent communication flows

Agent-Safe Pipeline

Repository: https://github.com/decionis/agent-safe-pipeline · Docs: https://decionis.com/docs Stars: 688 | Language: TypeScript | Primitive: independent policy-verdict boundary + single-use intent-bound grants Last pushed: 2026-10-13 | Last checked: 2026-10-15 (687→688★, +1 — last cycle’s +98 spike flattened; description remains the installable execution-authority gateway)

Agent-Safe Pipeline is the highest-traction new entrant since Bindu: created 2026-08-13 and already at 469★ with 59 forks as of 2026-08-16. Its README tagline — “Let agents propose. Let policy decide.” — frames the same gap as APS, decern, and Chancery, but as a service-boundary reference architecture rather than a kernel or IdP: Agent -> immutable intent -> Decionis -> ALLOW / ESCALATE / BLOCK -> SafeExecutor -> API, with a Presence service for verified human approval followed by Decionis re-evaluation. “Agents can reason, plan, and propose actions. They must not determine whether their own actions are authorized, possess downstream privileged credentials, or choose which trusted handler runs.”

Evidence checked (2026-08-16)

  • GitHub API: decionis/agent-safe-pipeline has 469★, 59 forks, 3 open issues, TypeScript as primary language, Apache-2.0 license, created 2026-08-13, pushed 2026-08-16; owner is an Organization; homepage https://decionis.com/docs.
  • README: six production invariants including canonical intent hashing with quick expiry, fail-closed behavior on network errors/missing grants/binding mismatches, grants bound to intent+decision+audience+expiry and consumed atomically before the handler runs, and downstream credentials existing only behind the trusted executor.
  • README/repo map: ships ARCHITECTURE.md, THREAT-MODEL.md, conformance canonical-hash test vectors (Unicode/astral, NFC vs NFD, negative zero, UTF-16 key sort order), SECURITY-EVIDENCE.md (control-to-artifact evidence map), FIXTURE-PROVENANCE.md, examples including a stdio MCP tool-gate and a Shopify refund agent, plus CI/CodeQL/OpenSSF Scorecard badges.
  • Structural note: Decionis and Presence remain closed hosted services (“the authoritative decision service… server internals can evolve independently behind versioned contracts”); the open repo is the client-side library and reference implementation.

Archon comparison

  • Direct overlap: pre-execution authorization, intent/delegation binding, fail-closed execution gating, single-use grants (compare Archon capability narrowing and receipts), MCP tool enforcement, tamper-evidence packaging
  • Agent-Safe Pipeline advantage: the strongest first-week traction ever observed in this landscape and the most complete evidence-packaging discipline (threat model + conformance vectors + security-evidence map) — this is what “production-shaped” looks like to enterprise buyers
  • Archon advantage: Agent-Safe Pipeline has no DID root, no credential issuance, no registry/discovery, and its root of decision authority is a closed hosted service; did:cid + VCs could supply the portable identity/authority layer its intents and grants bind to
  • Recommended stance: watch the star velocity past the launch window and treat as the new authorization-boundary benchmark alongside APS and decern. Bridge framing is natural: Archon credentials and delegation chains as the portable authority graph; a Decionis-class boundary as one enforcement point. The structural counter is the same as MolTrust’s — verdicts rented from one operator’s service vs. verifier-independent evidence

AgenticMail

Repository: https://github.com/agenticmail/agenticmail Stars: 237 | Language: TypeScript | Identity: Email-based / transport identity, no DID method Last pushed: 2026-10-14 | Last checked: 2026-10-15 (236→237★)

AgenticMail continues to grow and still describes itself as email, SMS, and phone-call infrastructure for AI agents. This is not a DID competitor, but it remains one of the strongest practical adoption signals in the agent infrastructure space.

Key signals

  • Strongest communications/transport traction signal in the tracked set
  • Real-world communications beat abstract identity in immediate operator legibility
  • Phone-call support broadens it from messaging transport into general real-world agent I/O

Archon comparison

  • Adjacent, not a like-for-like identity competitor
  • Strong integration target: DID-backed signatures, verifiable agent provenance, credential exchange, and trust metadata over real-world transport rails

MolTrust

Website: https://moltrust.ch | Operator: CryptoKRI GmbH, Zürich, Switzerland Stars: N/A (commercial service, not repo-centric) | Primitive: did:moltrust (draft method) + did:web + ERC-8004, Base L2 anchoring Last checked: 2026-10-15

MolTrust is a production commercial trust-infrastructure service for AI agents: W3C DID/VC identity, behavioral trust scoring, mandate enforcement, and on-chain audit anchoring sold as a hosted API with x402 micropayments. It surfaced through its moltrust-agent account on Moltbook, which promotes the service in agent-identity threads. The entry below is based on direct reads of the live API and published specifications, not secondhand claims.

Evidence checked (2026-07-21)

  • Live API: api.moltrust.ch reports v2.5 with /health returning {"status":"ok","database":"connected"}; the public no-key trust-score endpoint returns signed score payloads with per-component breakdowns; did:web:api.moltrust.ch resolves via /.well-known/did.json with Ed25519 verification methods.
  • did-method-spec (v0.1, April 2026): did:moltrust:<16-hex> derived from SHA-256 of the agent’s Ed25519 public key; key rotation with retained revoked-key history; Base L2 Merkle-batch anchoring of registrations; cross-ecosystem DID bridge (did:web, did:agentnexus, did:meeet) with trust-score import at 0.3 weight and 45-day half-life.
  • Product surface: 44+ MCP tools across 8 verticals (skills, prediction markets, shopping, travel, sports, music, sales, global); Agent Authorization Envelope (AAE) published as IETF Internet-Draft draft-kroehl-agentic-trust-aae-00; Interaction Proof Records (IPR) Merkle-batched and anchored on Base; cascade revocation across 8-hop delegation chains with CAEP events; SPIFFE bridge for enterprise workload identity.
  • Compliance: API endpoints mapped to EU AI Act Article 12 logging (with Article 50 transparency and Article 73 incident reporting), NIST AI RMF mapping, Singapore IMDA agentic-AI framework alignment; Circle Alliance membership.
  • Publications: arXiv:2605.06738 (“Trust Without Trusting — A Recomputable Trust Protocol for Autonomous Agents”), protocol tech spec v0.9, KYA whitepaper, sybil-resistance methodology — all hash-anchored on Base L2.
  • Pricing: free verification endpoints; $5 USDC per VC issuance via x402 on Base; subscriptions $19/mo (2 agents) to $299/mo (75 agents); Bitcoin Lightning listed as “coming soon” (PhoenixD prepped, not settling).
  • Blog: an AIP conformance comparison (claims full AIP feature coverage plus operational layer), and a June 2026 source-level analysis of Hermes Agent’s skill trust model (correctly identifies the four-repo TRUSTED_REPOS allowlist and the opt-in agent-created skill gate).
  • Re-checked 2026-08-16: /health still v2.5/ok; did:web:api.moltrust.ch still resolves with Ed25519 keys; pricing unchanged ($19–$299/mo, Lightning still “PhoenixD integration prepped, not yet settling”); homepage meta description now leads with “The trust layer for the agent economy” and closes with “Trust, by evidence.”; the AAE Internet-Draft advanced to draft-kroehl-agentic-trust-aae-01 (2026-08-11); the public repo now resolves as MoltyCel/moltrust-api (org capitalization changed from moltycel; 3★, pushed 2026-08-15).
  • Re-checked 2026-08-30: /health still v2.5/ok; did:web still resolves; homepage title/meta unchanged. The bare /pricing path now 301-redirects to /pricing.html (HTTP 200) — last week’s 403 was path-specific, not a takedown. Pricing re-verified: $19/mo (2 agents) to $299/mo (75 agents), $9/mo per additional agent, Lightning still “PhoenixD integration prepped, not yet settling.”
  • Re-checked 2026-09-06: /health still v2.5/ok; did:web still resolves; homepage title/meta unchanged. The /pricing path behavior changed again: it now 301-redirects to /pricing/, which returns 403 to curl even with a browser UA; /pricing.html still returns 200 and pricing was re-verified there ($19/mo (2 agents) to $299/mo (75 agents), $9/mo per additional agent, Lightning still “PhoenixD integration prepped, not yet settling”).
  • Re-checked 2026-09-13: /health still v2.5/ok; did:web still resolves; homepage title/meta unchanged. The /pricing path behavior is unchanged from last week (301→/pricing/→403 even with a browser UA); /pricing.html still returns 200 and pricing was re-verified there ($19/mo (2 agents) to $299/mo (75 agents), $9/mo per additional agent, Lightning still “PhoenixD integration prepped, not yet settling”). The AAE Internet-Draft advanced to draft-kroehl-agentic-trust-aae-02 (IETF Datatracker dated 2026-09-06).
  • Re-checked 2026-09-27: /health still v2.5/ok; did:web still resolves; homepage title/meta unchanged (“The Trust Layer for the Agent Economy”). /pricing path behavior unchanged (301→/pricing/→403); /pricing.html still returns 200 and pricing was re-verified there ($19/mo (2 agents) to $299/mo (75 agents), $9/mo per additional agent, Lightning still “PhoenixD integration prepped, not yet settling”). AAE draft unchanged at -02.
  • Re-checked 2026-10-11: /health still v2.5/ok (timestamp 2026-10-11 13:03 UTC); did:web still resolves; homepage title/meta unchanged. /pricing path behavior unchanged (301→/pricing/→403); /pricing.html still returns 200 and pricing was re-verified there ($19/mo (2 agents) to $299/mo (75 agents), $9/mo per additional agent, Lightning still “PhoenixD integration prepped, not yet settling”). AAE draft unchanged at -02 — third consecutive quiet live-surface cycle. Its public repo MoltyCel/moltrust-api pushed today (2026-10-11) and ticked 3→4★.
  • Re-checked 2026-10-15: /health still v2.5/ok (timestamp 2026-10-15 17:24 UTC); did:web still resolves; homepage title/meta unchanged. /pricing path behavior unchanged (301→/pricing/→403); /pricing.html still returns 200 and pricing was re-verified there ($19/mo (2 agents) to $299/mo (75 agents), $9/mo per additional agent, Lightning still “PhoenixD integration prepped, not yet settling”). AAE draft unchanged at -02 — fourth consecutive quiet live-surface cycle. Its public repo MoltyCel/moltrust-api pushed 2026-10-15 (flat at 4★).

Architecture / trust model

  • Centralized commercial registry: did:moltrust resolution routes through api.moltrust.ch; trust scores (0–100) are operator-issued from an endorsement graph, interaction history, cross-vertical coverage, and sybil penalties. Cold-start scores derive from on-chain/ERC-8004 data.
  • Blockchain anchoring on Base L2 for registrations, violation records, and spec versions — anchoring provides tamper-evidence, but the root of resolution and scoring remains the operator.
  • MoltProof (read-only mandate-verdict engine) and MoltGuard (prediction-market anomaly scanning) are genuinely public-falsifiable components: verdicts recomputable from public chain data.

Archon comparison

  • Direct overlap: agent DID, VC issuance/verification, delegation chains with revocation, receipts/audit anchoring, MCP tooling, EU AI Act compliance positioning, agent-commerce payments.
  • MolTrust advantage: production maturity and breadth today — live API, shipped verticals, SDKs for npm/CrewAI/LangChain/OpenClaw, offline verification package (@moltrust/verify), SLA tiers, and aggressive protocol-agnostic outreach (AIP conformance docs, OpenClaw plugin, Hermes skill-trust analysis).
  • Archon advantage: sovereign did:cid root authority vs. operator-issued identity; decentralized hyperswarm registry vs. resolution that depends on one company’s API; multiparty Gatekeeper/Keymaster separation with no single point of control; Lightning-native settlement adjacency while MolTrust’s Lightning rail is roadmap-only.
  • Recommended stance: treat MolTrust as the clearest “centralized, done well” benchmark and the strongest current narrative foil. The structural counter is reputation-as-a-service from a single operator vs. evidence-first, verifier-independent receipts. It is also a credible bridge target: did:cid as portable root authority with MolTrust-style operational layers (scoring, compliance, vertical VCs) consuming it rather than replacing it.

Grantex

Repository: https://github.com/mishrasanjeev/grantex Stars: 33 | Language: TypeScript | Primitive: delegated authorization / Commerce Passport / audit Last pushed: 2026-10-15 | Last checked: 2026-10-15 (flat at 33★)

Grantex is a new high-signal entrant. Its repo description frames it as identity, authorization, and audit infrastructure for AI agents — the “OAuth moment” for the agentic internet. The README frames it as a delegated authorization protocol for AI agents and a commerce consent/passport/policy/audit/payment-control layer.

Key signals

  • Directly attacks the authorization gap rather than only identifier creation
  • Strong commerce framing: scoped consent, audit, policy, payment-control
  • README explicitly says production Commerce V1 discovery is currently disabled/fail-closed, so this is not yet proven production checkout infrastructure

Archon comparison

  • Not a DID-method competitor; it is an authorization/consent/commercial-action layer
  • Strategically important because agent identity is converging with delegated authority and payments
  • Archon should be able to issue, verify, or anchor the identity side of a Grantex-like authorization flow

Agent Passport System

Repository: https://github.com/agent-passport-system/agent-passport-system (moved from aeoess/) Stars: 49 | Language: TypeScript | Primitive: did:aps, accepted external identities, monotonic delegation, signed receipts Last pushed: 2026-10-15 | Last checked: 2026-10-15 (flat at 49★; repos under the agent-passport-system org; IETF draft-pidlisnyi-aps still -04, 2026-09-28)

Companion repos under aeoess/ observed pushing 2026-08-01: agent-passport-python, agent-passport-go, agent-passport-mcp. On 2026-08-09 the byte-level conformance suite had moved to a dedicated org — Agent-Authority-Conformance/aps-conformance-suite (1★, pushed 2026-08-13) — alongside a new Agent-Authority-Conformance/governance repo (0★, pushed 2026-08-06); aeoess/aps-web now returns 404. APS is positioning as a multi-repo verifiable protocol with its conformance surface breaking out into its own organization, not a single repo.

On 2026-08-16 the SDK family grew again: aeoess/agent-passport-rust (0★, created 2026-08-15) is a verification-only Rust implementation of APS artifact verifiers — passports, delegations, chains, ReceiptV1, RFC 8785 and legacy canonicalization — with no key generation, signing, or issuance. Its README references Internet-Draft draft-pidlisnyi-aps-03 (“Agent Passport System (APS): Verifiable Agent Identity, Faceted Authority, and Signed Action Receipts”), an individual submission now at revision -03. Verification-only SDKs are how a protocol recruits third-party verifiers without surrendering issuance.

Agent Passport System is the clearest new direct pressure on Archon’s delegated authority + receipt story. Its README describes it as an enforcement and accountability layer for AI agents that accepts did:key, did:web, SPIFFE SVIDs, OAuth tokens, and native did:aps. The core claim is not just identity: authority can only decrease at each transfer point, gateway enforcement happens before execution, and every action produces a signed receipt.

Evidence checked

  • GitHub API: aeoess/agent-passport-system has 24★, 11 forks, 9 open issues/PRs, TypeScript as primary language, Apache-2.0 license, and was pushed 2026-07-05.
  • README: positions the project as “enforcement and accountability layer for AI agents” and “bring your own identity.”
  • README: says it accepts did:key, did:web, SPIFFE SVIDs, OAuth tokens, and native did:aps.
  • README: describes monotonic narrowing, gateway enforcement, a 3-signature action chain, commerce preflight, Bayesian reputation, and signed receipts for allow/deny outcomes.

Archon comparison

  • Direct overlap: agent identity, scoped delegation, capability boundaries, action receipts, payment/commerce guardrails, reputation/audit framing
  • Agent Passport advantage: very crisp authority-narrowing and receipt language; BYO identity stance makes it easier to bridge across existing enterprise/auth ecosystems
  • Archon advantage: sovereign root identity via did:cid, decentralized registry/discovery, VC/status-list orientation, Dmail/Lightning adjacency, and stronger substrate independence
  • Recommended stance: treat APS as a serious authority-layer benchmark and likely bridge target. Archon should be able to issue or verify authority that APS-style gateways enforce, while preserving did:cid as the root identity rather than adopting did:aps as the anchor

decern

Repository: https://github.com/anivar/decern Stars: 13 | Language: Rust | Primitive: AuthZEN-shaped PDP + cvc5 SMT-proven safety invariants + signed hash-chained decision ledger Last pushed: 2026-10-14 | Last checked: 2026-10-15 (flat at 13★)

decern is the sharpest new authorization-layer entrant this cycle: created 2026-08-02 and already at 12★ with 3 forks. Its README frames the gap directly — “The industry standardizes how authority is represented … and defers the guarantee (that attenuation holds, that nothing was dropped from the log, that a decision stayed within its mandate) to implementer policy. decern is the guarantee.” The kernel treats humans, agents, and workloads as one principal type; a decision is a pure function of (principal, authority graph, policy, now); and 9 named invariants (money-gate, isolation, decay, attenuation-edge, scope-gate, revocation-gate, residency-gate, role-gate, consent-gate) are discharged by the cvc5 SMT solver across the entire input space rather than sampled by tests. Every decision lands in a hash-chained, signed ledger that any third party can verify offline with decern verify.

Evidence checked (2026-08-09)

  • GitHub API: anivar/decern has 12★, 3 forks, Rust as primary language, Apache-2.0 license, created 2026-08-02, pushed 2026-08-09.
  • README: two binaries (decern to prove/verify, decern-serve as PDP); AuthZEN-shaped /access/v1/evaluation HTTP API; thin clients published for Python (uv add decern), TypeScript (npm install decern), and Go; prebuilt signed binaries for Linux/macOS/Windows; Zenodo DOI (10.5281/zenodo.21848620).

Archon comparison

  • Direct overlap: delegated authorization, attenuation/monotonic narrowing, revocation, tamper-evident decision logs, offline verification — the same “prove this agent was allowed to do X” surface as APS and Chancery
  • decern advantage: formal-methods posture (SMT-checked invariants) is a stronger correctness story than test suites, and the README language is the most precise statement of the authorization-guarantee gap in the tracked set
  • Archon advantage: decern has no DID root, no credential issuance, no registry/discovery — principals are local identifiers. did:cid + VCs could supply the portable identity/authority graph a decern-style PDP decides over
  • Recommended stance: watch closely and treat as the new authorization-kernel benchmark alongside APS. Bridge framing is natural: Archon issues the credentials and delegation chain; a decern-class PDP enforces and logs the decision

SIQ Agent Security

Repository: https://github.com/maoyadongsh/siq-agent-security Stars: 50 | Language: Python | Primitive: Ed25519-signed intent/parameter/receipt binding + “Skill Execution Context” permission boundary Created: 2026-08-13 | Last pushed: 2026-10-09 | Last checked: 2026-10-15 (51→50★, -1)

SIQ Agent Security is the highest-traction discovery-sweep entrant since Agent-Safe Pipeline: 51★ / 12 forks, Apache-2.0, signed 0.4.0 release shipped today. Its README (bilingual, Chinese-primary) frames the project as a “secure runtime for agent skills”: a local-first authorization layer that discovers an operator’s agents and installed skills, tracks permissions, gates execution through host hooks, and binds trusted authorization intent, action parameters, and execution receipts together with Ed25519 signatures and SHA-256 content digests so that source authenticity, content integrity, and authorization consistency can each be independently verified. It ships both a personal client (local service + browser console) and an enterprise variant (control plane, Edge, connectors, periodic discovery, audit export) from the same codebase, though the enterprise business-identity federation path is explicitly marked unverified in the README.

Evidence checked (2026-09-27)

  • GitHub API: maoyadongsh/siq-agent-security has 51★, 12 forks, Apache-2.0 license, Python as primary language, created 2026-08-13, pushed 2026-09-27.
  • README: signed release 0.4.0 pinned to commit 2cd6116; ships CI and “research reproduction” GitHub Actions badges; explicitly states research demos run deterministic fixtures on plain Linux without model API keys or GPU.
  • README: distinguishes “asset discovery” from “managed” from “protected” — discovery does not imply enforcement is actually active; results are only claimed once effect-capture evidence and independent verification exist.

Archon comparison

  • Direct overlap: pre-execution authorization intent binding, signed execution receipts, tamper-evident verification chain, skill/tool permission boundary enforcement — the same “prove this agent was allowed to do X” surface as APS, decern, and Agent-Safe Pipeline
  • SIQ advantage: fastest-growing authorization-boundary repo since Agent-Safe Pipeline’s launch week; ships both individual and enterprise control-plane variants from one codebase; unusually explicit about what is verified vs. merely shipped
  • Archon advantage: SIQ has no DID root, no credential issuance, no decentralized registry — it authorizes local/enterprise skill execution, not portable agent identity. did:cid + VCs could anchor the identity side of a SIQ-style authorization intent
  • Recommended stance: add to the watchlist and track star velocity next cycle; bridge framing is the same as Agent-Safe Pipeline and decern — Archon issues portable identity/credentials, a SIQ-class runtime enforces and logs the decision locally

AgentNexus

Repository: https://github.com/kevinkaylie/AgentNexus Stars: 10 | Language: Python | Primitive: DID + relay + encrypted messaging + capability tokens Last pushed: 2026-09-23 | Last checked: 2026-10-15 (flat at 10★)

AgentNexus is a collaboration substrate for agent teams. Its README says the public positioning has converged on DID identity, authorization, artifact delivery, and objective loops for heterogeneous agents. It began as an agent “WeChat / WhatsApp” idea, but the current docs frame a broader workflow layer: DID identity, relay, encrypted messages, access control, vault, playbook, context snapshots, handoff checkpoints, delivery manifests, and owner takeover.

Evidence checked

  • GitHub API: kevinkaylie/AgentNexus has 9★, Python as primary language, and was pushed 2026-07-02.
  • README: says AgentNexus provides identity, communication, discovery, authorization, shared artifacts, and team workflow orchestration for AI agents.
  • README: describes an Objective Loop using Worker DIDs, artifact handoff, acceptance receipts, and secretary/human interaction.

Archon comparison

  • Direct overlap: DID identity, authorization, agent communication, artifact handoff, receipt/acceptance semantics, multi-agent workflow coordination
  • AgentNexus advantage: clearer team-workflow and artifact-handoff product surface
  • Archon advantage: decentralized identity root, credential architecture, registry/discovery substrate, payment-aware settlement, and portable authority independent of a specific team workflow system
  • Recommended stance: watch and bridge. AgentNexus-like workflows are a natural consumer of Archon credentials, Dmail, Lightning settlement, and signed work receipts

Kestrel Sovereign

Repository: https://github.com/KestrelSovereignAI/kestrel-sovereign Stars: 8 | Language: Python | Primitive: portable DID identity + signed constitutional governance Last pushed: 2026-10-15 | Last checked: 2026-10-15

Kestrel is not mainly a DID protocol; it is a sovereign-agent framework. The README describes agents with portable user-owned DID identity, local-first persistent memory, and constitutional governance where amendments require cryptographic signature. That overlaps Archon’s sovereignty narrative because it sells the whole agent as portable and user-owned, not just a credential layer.

Evidence checked

  • GitHub API: KestrelSovereignAI/kestrel-sovereign has 7★, 2 forks, 40 open issues/PRs, Apache-2.0 license, Python as primary language, and was pushed 2026-07-05.
  • README: says every deployed agent is owned by its user, governed by immutable principles, and able to remember across conversations.
  • README: names portable DID identity, persistent memory, and constitutional governance as the three pillars.

Archon comparison

  • Direct overlap: sovereign agent identity, user ownership, continuity across providers, cryptographic governance
  • Kestrel advantage: full agent framework narrative with memory and governance already packaged
  • Archon advantage: dedicated DID/VC authority substrate, registry/discovery, credential issuance/verification, external service endpoints, payment/receipt adjacency
  • Recommended stance: classify as adjacent framework pressure. Archon should be the identity/authority rail a Kestrel-like agent can carry across runtimes, not another full agent framework competing feature-for-feature

Agentic Airlock

Repository: https://github.com/airlock-protocol/airlock Stars: 2 | Language: Python | Primitive: Ed25519 identity verification + OAuth 2.1 tokens + delegation chains Last pushed: 2026-09-25 | Last checked: 2026-10-15

Agentic Airlock is an OAuth/compliance-oriented trust layer. Its README frames the project as a trust and compliance layer for AI agents, extending OAuth 2.1 with progressive trust, delegation chains, and tamper-evident audit trails. It explicitly positions the gap as agent communication protocols such as A2A and MCP lacking standard identity, authorization, and trust verification.

Evidence checked

  • GitHub API: airlock-protocol/airlock has 2★, 11 open issues/PRs, Apache-2.0 license, Python as primary language, and was pushed 2026-07-05.
  • README: lists OAuth 2.1 authorization server support, Ed25519 private_key_jwt, EdDSA-signed JWT access tokens with trust score claims, RFC 8693 token exchange, introspection, OIDC discovery, compliance records, and hash-chain integrity.
  • README: says it accepts both Ed25519 signatures and OAuth bearer tokens.

Archon comparison

  • Direct overlap: agent identity, delegation, trust verification, audit trail, compliance positioning
  • Airlock advantage: familiar OAuth/OIDC framing and regulatory/audit language
  • Archon advantage: sovereign DID root, portable VC authority, decentralized registry/discovery, and payment/receipt composability without requiring an OAuth authority to be the identity root
  • Recommended stance: treat as enterprise/auth-stack pressure. Archon should explain when DID/VC credentials can replace OAuth dependency and when bridge-to-OAuth is the practical integration path

Chancery

Repository: https://github.com/chanceryhq/chancery Stars: 26 | Language: Go | Primitive: registry + scoped delegation + in-path MCP enforcement Last pushed: 2026-07-21 | Last checked: 2026-10-15

Chancery is a new identity-provider-shaped entrant for AI agents. Its GitHub metadata describes a registry, scoped delegation, in-path MCP enforcement, instant revocation, and tamper-evident audit. Traction jumped 0→25★ between 2026-07-11 and 2026-08-02 — the sharpest relative move in the tracked set — but it has since stalled on pushes: no commit since 2026-07-21 through the 2026-08-09, 2026-08-16, 2026-08-23, 2026-08-30, 2026-09-06, 2026-09-13, 2026-09-27, 2026-10-04, 2026-10-11, and 2026-10-15 sweeps (ten consecutive stalled-push cycles), even though its star count ticked 25→26 on 2026-09-27 with no accompanying push. As of 2026-09-06 it is demoted from watch to inactive at the current evidence level. It remains strategically relevant because it names the exact enterprise control-plane pressure that can make identity practical: enforcement at the tool boundary, not just identifier issuance.

Archon comparison

  • Direct overlap: scoped delegation, revocation, audit, and MCP/tool enforcement
  • Chancery advantage: crisp IdP/control-plane framing for enterprise operators
  • Archon advantage: sovereign did:cid root authority, decentralized registry/discovery, W3C VC/status-list orientation, Dmail/Lightning adjacency, and independence from a single hosted IdP
  • Recommended stance: treat Chancery as an enterprise enforcement benchmark. Archon should show how DID/VC capabilities can drive MCP enforcement while keeping the root identity decentralized

AgentValet

Repository: https://github.com/AgentValet/AgentValet Stars: 2 | Language: TypeScript | Primitive: IETF AIMS + SPIFFE + AuthZEN + CIBA + MCP proxy Last pushed: 2026-09-26 | Last checked: 2026-10-15 (flat at 2★)

AgentValet frames itself as IGA for AI agents: identity, credential governance, and an MCP proxy using enterprise-native language such as IETF AIMS, SPIFFE, RFC 7591, AuthZEN, and CIBA. It is early, but it matters because it translates the agent-authority problem into the governance vocabulary that security teams already use.

Archon comparison

  • Direct overlap: agent identity governance, credential governance, MCP proxy enforcement, and enterprise authorization patterns
  • AgentValet advantage: standards/integration vocabulary familiar to enterprise IAM teams
  • Archon advantage: decentralized identity root, portable credentials/capabilities, service endpoints, and payment/receipt composability outside a classic IGA control plane
  • Recommended stance: bridge rather than dismiss. Archon should be able to export credentials or authority proofs that enterprise IGA/MCP proxies can consume while preserving did:cid as root authority

Agent Name Service (ANS)

Repository: https://github.com/agentnameservice/ans (+ org: ans-registry, Go/Rust/Java SDKs, agent-trust-discovery) Stars: 43 (ans) / 33 (ans-registry) | Language: Go | Primitive: IETF-draft naming registry (draft-narajala-ans-00) + append-only Merkle transparency log + SCITT COSE_Sign1 receipts Last pushed: 2026-10-15 / 2026-10-15 | Last checked: 2026-10-15

ANS is an open-source implementation of the Agent Name Service IETF draft: a registry plus transparency log for discovering and verifying AI agents by name. Every registered agent gets a versioned DNS-style name (ans://v1.0.0.my-agent.example.com), a publicly auditable append-only Merkle event history, SCITT COSE_Sign1 receipts proving point-in-time agent state, identity certificates signed by a private CA for agent mTLS, optional BYOC server certs with pinned TLSA records, and an offline CLI verifier (ans-verify). The companion agent-trust-discovery repo computes a per-agent Trust Vector across five spec dimensions (integrity, identity, solvency, behavior, safety) with explainable scoring; its make demo-live pulls a live production registry snapshot described as GoDaddy’s ANS deployment.

Archon comparison

  • Direct overlap: agent registry/discovery, tamper-evident state, receipts, verification tooling
  • ANS advantage: IETF-draft legitimacy, DNS/CA operational familiarity, multi-language SDKs, and a visible production deployment
  • Archon advantage: ANS is CA/DNS-anchored — the root of naming authority is still a certificate hierarchy and DNS control, not a sovereign content-addressed identity. did:cid + the Hyperswarm registry has no single point of control; ANS receipts prove state about a name, not self-sovereign authority over it
  • Recommended stance: watch whether the GoDaddy-affiliated deployment gains outside adoption. Bridge framing is available — ANS-style transparency-log receipts could attest state about did:cid roots — but do not let DNS/CA naming become the root identity story

Attestix

Repository: https://github.com/VibeTensor/attestix Stars: 874 | Language: Python | DID Method: did:key / did:web Last pushed: 2026-10-15 | Last checked: 2026-10-15 (flat at 874★; resumed commits after a pause since 2026-10-01 — the +857 spike is still not compounding; unverified-traction read stands)

Attestix remains a compliance-forward attestation stack. Current GitHub metadata describes DID-based agent identity, W3C Verifiable Credentials, EU AI Act compliance, delegation chains, reputation scoring, and 47 MCP tools across 9 modules.

Key features / signals

  • Compliance wedge remains clear and increasingly concrete
  • Continued activity since the April sweep
  • MCP tooling count makes the project more operationally legible than a pure spec

Archon comparison

  • Best viewed as a complementary compliance layer
  • Partnership/integration candidate: Archon supplies sovereign identity + credential substrate; Attestix supplies compliance workflows and reputation/compliance packaging

Hedera / did:hedera

Primary sources: https://github.com/hashgraph/did-method · https://github.com/hashgraph/did-sdk-java · https://docs.hedera.com/solutions/ai/index.md · https://docs.hedera.com/solutions/ai/hosted-mcp-server.md · https://docs.hedera.com/solutions/ai/x402.md Signals checked: did-method 27★, did-sdk-java 36★, hedera-agent-kit-js 69★ (pushed 2026-10-14) | Last checked: 2026-10-15

Hedera belongs in the report twice: as a direct DID-method competitor and as a broader enterprise agent substrate. The hashgraph/did-method repository describes the Hedera DID method specification; the spec identifies the namestring as hedera and requires DIDs to begin with did:hedera. The Java SDK repository says it supports Hedera Hashgraph DID Method and Verifiable Credentials using Hedera Consensus Service.

On the adjacent-substrate side, Hedera now has explicit AI-agent positioning: AI Studio for verifiable AI agents, Agent Kit for LLM-powered Hedera network operations, a Hosted MCP Server for exposing Agent Kit tools to MCP clients, x402 payments for HBAR/HTS inside HTTP requests, and HCS consensus timestamps for audit logs.

Key signals

  • did:hedera is a registered DID-method-style competitor to did:cid, not just generic blockchain infrastructure
  • DID/VC SDK work exists, but the primary Hashgraph DID repositories show modest traction and slower direct DID activity than Hedera’s newer agent/payment push
  • HCS provides a clear audit-log substrate for agent actions, receipts, and compliance proofs
  • x402 on Hedera overlaps with Archon’s payment-aware authority and agent-commerce direction
  • Enterprise governance/compliance positioning is much stronger than most small agent-identity repos

Archon comparison

  • Direct overlap: DID method, DID document lifecycle, VC support, service endpoints, registry/resolution semantics
  • Architectural split: Hedera anchors identity operations to the Hedera/HCS network; Archon emphasizes content-addressed identity, Hyperswarm registry/discovery, and stronger substrate independence
  • Threat: medium as a direct DID competitor, high as an enterprise settlement/audit/payment substrate
  • Strategy: bridge where useful, compete on sovereignty. Archon should be able to explain why did:cid is the sovereign root of authority while still integrating with Hedera for optional settlement, HCS audit anchoring, or x402-compatible paid APIs

Soulverse

Website: https://www.soulverse.world/ Observed public surface: live site, Indicio announcement, and live-presentation report; no relevant public SDK repo/package observed Last checked: 2026-10-15

Soulverse is a new adjacent pressure point because its public site positions the platform as pre-execution validation infrastructure for global identity and trust. The homepage says Soulverse resolves identity, credentials, authority, governance, rules, and settlement into a deterministic layer so cross-boundary actions are validated before execution. Its solutions page explicitly includes Agentic Validation for autonomous AI systems: intent objects, model integrity attestations, capability envelopes, live policy recalculation, and step-level authorization for tool invocation.

The strongest Archon overlap is not plain DID issuance; it is the agent-action control path. Soulverse says its Soul AI Agent SDK enables credential-gated autonomous actions, policy-scoped execution limits, and audit-ready action logging, with agents verifying identity and authority before every action. That makes it relevant to Archon’s authority/delegation/receipt story even though Soulverse does not currently show the same Lightning-native payment-settlement wedge.

Evidence checked

  • Live homepage: title Soulverse | The Operating System of Trust; description says global identity and trust infrastructure with sovereign identity as the root layer for institutional systems.
  • Live homepage: describes three layers: Identity Layer, Trust Layer, and Settlement Layer, including W3C DIDs, SSI wallet, biometric binding, verifiable credentials, runtime verification, identity-bound settlement, cross-chain finality proofs, and audit trail.
  • Live technical brief: lists six validation domains: identity state, credential status, relational trust models, policy thresholds, enforcement conditions, and settlement requirements.
  • Live solutions page: describes Agentic Validation for autonomous AI systems, including intent binding, model integrity verification, capability envelope enforcement, live policy recalculation, and step-level authorization.
  • Live Product SDKs page: claims Soul ID SDK, Trust Protocol SDK, Trust Engine SDK, Credential SDK, Soulwrapper SDK, Soulbridge SDK, Soulogram SDK, Soul AI Agent SDK, Chrome Extension SDK, and Soul Super Wallet SDK.
  • npm checks on 2026-07-15: @soulverse/soul-id-sdk, @soulverse/trust-protocol-sdk, @soulverse/soul-ai-agent-sdk, @soulverse/credential-sdk, and @soulverse/soulbridge-sdk returned 404.
  • npm re-check on 2026-08-16: @soulverse/soul-id-sdk, @soulverse/trust-protocol-sdk, and @soulverse/soul-ai-agent-sdk all still return 404; homepage title unchanged (Soulverse | The Operating System of Trust).
  • npm re-check on 2026-08-23: all three packages still 404; homepage title still unchanged.
  • npm re-check on 2026-08-30: all three packages still 404 (sixth consecutive weekly check); homepage title still unchanged.
  • npm re-check on 2026-09-06: all three packages still 404 (seventh consecutive weekly check); homepage title still unchanged.
  • npm re-check on 2026-09-13: all three packages still 404 (eighth consecutive weekly check); homepage title still unchanged.
  • npm re-check on 2026-09-27: all three packages still 404 (ninth consecutive weekly check); homepage title still unchanged.
  • npm re-check on 2026-10-11: all three packages still 404 (eleventh consecutive weekly check); homepage title still unchanged.
  • npm re-check on 2026-10-15: all three packages still 404 (twelfth consecutive weekly check); @nexionplatform/sdk still 404; homepage title still unchanged.
  • GitHub repository searches on 2026-07-15 for "Soulverse" "Soul AI Agent SDK", "@soulverse/soul-id-sdk", and "did:soul" Soulverse returned no relevant public repos.
  • Indicio announcement, 2025-04-30: Indicio welcomed Soulverse as a Network Partner; the announcement says Soulverse would use Indicio MainNet for identity-ledger transactions and describes Soul Super Wallet as combining verifiable credentials, digital assets, fiat currency, and biometric authentication.
  • Live-presentation report from Cypher, 2026-07-15: the did:soul team said they are considering decentralization in the future. This is treated as evidence that current did:soul decentralization is not implemented or externally verifiable.

Archon comparison

  • Direct overlap: identity, credentials, authority, governance, execution gating, agent policy enforcement, audit-ready action logging, and settlement/audit language
  • Current identity-root caveat: did:soul appears proprietary / Soulverse-controlled today; the team reportedly frames decentralization as a future consideration rather than a present property
  • Soulverse advantage: polished institutional trust-infrastructure narrative, Indicio ecosystem signal, broad SDK/menu surface, and explicit agentic validation language
  • Archon advantage: sovereign did:cid root authority available now, decentralized registry/discovery, content-addressed identity, Dmail/Lightning adjacency, and clearer paid-task/receipt settlement wedge
  • Recommended stance: track Soulverse as an agent-governance and execution-gating watchlist entry; compete on decentralized root authority and Lightning-native settlement while treating Indicio/Soulverse-style VC infrastructure as a possible bridge surface

didit skills

Repository: https://github.com/didit-protocol/skills Stars: 28 | Language: Python | Scope: Identity verification / KYC API wrappers Last pushed: 2026-10-12 | Last checked: 2026-10-15

Official Didit agent skills for identity verification, KYC, AML screening, biometric APIs, and session management. This remains an adjacent identity-verification API wrapper set, not a decentralized agent identity competitor.

Archon comparison

  • Non-competitor; adjacent compliance/KYC infrastructure
  • Relevant if Archon agents need to bridge decentralized identity with existing verification rails

AIP (Agent Identity Protocol)

Repository: https://github.com/The-Nexus-Guard/aip Stars: 14 | Language: Python | DID Method: did:aip / Ed25519 Last pushed: 2026-03-22 | Last checked: 2026-10-15

Cryptographic identity, trust chains, and E2E encrypted messaging for AI agents. AIP’s repo description still emphasizes pip-installable identity, signed trust chains, and encrypted communications.

Key features / signals

  • Slight growth since April, but less active than the newest entrants
  • Trust-chain framing remains distinct from pure DID issuance stacks

Archon comparison

  • Partial overlap: stronger on social trust semantics; weaker on interoperable credential framing and decentralized registry architecture
  • Watch because trust-chain language is understandable to developers and operators

clawdentity

Repository: https://github.com/vrknetha/clawdentity Stars: 9 | Language: TypeScript | DID Method: did:cdi Default branch: develop | Last pushed: 2026-04-22 | Last checked: 2026-10-15

clawdentity still positions itself as the messaging layer for AI agents: any agent can DM or group-chat with any other agent across platforms.

Key features / signals

  • Closest philosophical neighbor to Archon in content-derived identity thinking
  • Messaging-first framing remains clearer than identity-substrate framing for many users
  • Development has slowed relative to May’s new entrants, but the concept remains strategically relevant

Archon comparison

  • clawdentity wins on messaging-first clarity
  • Archon differentiates on decentralized registry design, credential stack, and stronger sovereignty story
  • High-value compare/contrast target because the overlap is concrete

Motebit

Repository: https://github.com/motebit/motebit Stars: 7 | Language: TypeScript | Primitive: Ed25519 identity + signed execution receipts Last pushed: 2026-10-14 | Last checked: 2026-10-15 (flat at 7★)

Motebit describes itself as an open protocol and reference runtime for sovereign AI agents. Its README frames identity as persistent across devices/providers/time, trust as signed execution receipts, and governance as a fail-closed policy boundary.

Key signals

  • Very early, but philosophically close to Archon
  • Explicitly combines identity, memory, trust, and governance
  • Strong language around receipts and runtime boundaries

Archon comparison

  • Potential philosophical rival or integration target, depending on implementation depth
  • Archon should watch the “execution receipt” and “policy gate” framing closely; it may be easier for operators to grasp than DID semantics

Credat

Repository: https://github.com/credat/credat Stars: 2 | Language: TypeScript | Primitive: scoped credentials Last pushed: 2026-05-22 | Last checked: 2026-10-15

Credat frames itself as a trust layer for AI agents: identity, delegation, and mutual verification in a single TypeScript package. Its README emphasizes scoped owner-issued credentials, service-side verification, small bundle size, and tests.

Archon comparison

  • Directly relevant at the authorization/delegation layer
  • Less sovereign/decentralized than Archon, but the developer pitch is concise and practical
  • Useful benchmark for Archon’s SDK ergonomics and “prove this agent is allowed to act” story

HelixID

Repository: https://github.com/helixid/helixid (moved from dgverse-labs/helixid) Stars: 8 | Language: TypeScript | Primitive: DID + VC + scoped permissions Last pushed: 2026-09-29 | Last checked: 2026-10-15 (flat at 8★)

HelixID describes itself as an open-source identity and authorization layer for AI agents. Its README frames the gap as delegation chains, scoped authority, cross-org trust, revocation, and audit trail problems for API-key-driven agents.

Archon comparison

  • Standards-aligned watchlist item
  • Very low traction so far, but it names the same pain points Archon needs to address publicly

IDProva

Repository: https://github.com/techblaze-au/idprova Stars: 1 | Language: Rust | Primitive: Ed25519 keys + delegated authority + hash-chained audit receipts Last pushed: 2026-07-24 | Last checked: 2026-10-15

IDProva frames itself as cryptographic identity for AI agents, designed to sit alongside existing enterprise IdPs. Its README emphasizes questions like who an agent is, what it is allowed to do, who granted permission, and whether audit trails can be proven untampered.

Archon comparison

  • Enterprise-adjacent identity/audit layer rather than sovereign DID infrastructure
  • Worth watching because enterprise adoption may form around auditability and delegation before it cares about decentralized registries

A2AL

Repository: https://github.com/a2al/A2AL Stars: 1 | Language: Go | Primitive: cryptographic AID Last pushed: 2026-10-07 | Last checked: 2026-10-15

A2AL describes itself as an agent-to-agent networking protocol for publishing, discovering, and securely connecting agents without central infrastructure. It ships as a daemon with a built-in MCP server.

Archon comparison

  • Adjacent P2P discovery and communication layer
  • Could overlap with Archon’s decentralized registry/discovery story if it gains traction

Chorus

Repository: https://github.com/LyonMask/chorus Stars: 1 | Language: Rust | Primitive: decentralized identity + libp2p Last pushed: 2026-06-28 | Last checked: 2026-10-15

Chorus describes itself as the open communication layer for AI agents: peer-to-peer, end-to-end encrypted, and no central servers.

Archon comparison

  • Similar to A2AL: communication/discovery rather than identity substrate
  • Worth watching because P2P encrypted agent communication is a natural place to embed or require identity proofs

Repository: https://github.com/payelink/payelink-agent-identity-sdk Stars: 2 | Language: Python | DID Method: did:key Last pushed: 2026-02-09 | Last checked: 2026-10-15

Python SDK + CLI for minting and resolving did:key identifiers. Still identity-only: useful, standards-aware, but narrow in scope.

Archon comparison

  • Archon remains substantially broader: decentralized registry, credentials, and persistent agent identity narrative
  • payelink is best interpreted as an SDK component inside a larger stack rather than a standalone ecosystem anchor

agent-did

Repository: https://github.com/dantber/agent-did Stars: 0 | Language: TypeScript | DID Method: did:key Last pushed: 2026-02-06 | Last checked: 2026-10-15

A W3C-compliant DID and VC toolkit for AI agents, with credential issuance and scoped capabilities. It remains a clean minimal comparison point but has not gained traction.

Archon comparison

  • Closest direct DID/VC competitor in standards terms
  • Archon differentiates through decentralized discovery, content-addressed DID method, and stronger sovereign-agent framing

agent-identity-hub

Repository checked: https://github.com/yksanjo/agent-identity-hub Last checked: 2026-10-15

The repository still returns 404 / Not Found via the GitHub API. Earlier notes described it as a did:ethr-based swarm/orchestration layer, but it is not currently inspectable.

Archon comparison

  • Keep as a historical note only
  • Remove from active competitive pressure unless the repo reappears or moves

Competitive Matrix (2026-07-15 Snapshot)

Feature Archon Bindu Agent Passport System Chancery / AgentValet Urbit ANP / AgentConnect AgenticMail Grantex / Airlock Hedera / did:hedera Soulverse AgentNexus
Primary role DID + credential + registry stack Agent identity + A2A + auth + payments platform Enforcement + delegation + receipt layer Enterprise agent IdP / IGA / MCP enforcement Personal server OS + P2P network + identity substrate Agent communication protocol + SDK Real-world communications infra Delegated auth + OAuth/compliance audit DID method + agent/payment/audit substrate Agent governance + pre-execution validation Agent team communication/workflow substrate
Identity primitive did:cid did:bindu + Ed25519 + Hydra client metadata did:aps + BYO DID/SPIFFE/OAuth registry / AIMS / SPIFFE / AuthZEN / CIBA Urbit ID / Azimuth PKI did:wba Email/SMS/phone identity Commerce Passport / OAuth 2.1 / Ed25519 did:hedera + HCS did:soul proprietary / future decentralization under consideration + W3C DID/VC claims DID + relay identity
Registry / control plane Hyperswarm P2P, BTC:mainnet optional Bindu/Hydra auth + resolver + app platform Gateway / policy evaluator IdP, MCP proxy, revocation and governance plane Azimuth.eth + Urbit hierarchy + Ames/Jael Web-hosted DID + protocol/SDK Hosted transport infrastructure Cloud/service-oriented auth plane Hedera Consensus Service / Hedera network Trust Protocol / validation engine claims Relay / vault / playbook system
Truly decentralized ✅ ❌ / cryptographic but platform-centered ⚠️ BYO identity, gateway-centered enforcement ❌ / enterprise control-plane model ✅ / hierarchical + Ethereum-rooted ⚠️ web-federated ❌ ❌ / service-oriented ⚠️ public DLT with permissioned council governance ❌ / decentralization described as future consideration; public root not verified ⚠️ relay/federation model
Credential issuance ✅ W3C VC 2.0 + status lists ⚠️ DID signatures / skills / auth emphasis ⚠️ passports/delegations, not W3C VC core ✅ / governance emphasis, not sovereign DID root ❌ / not W3C VC-native ⚠️ identity/auth emphasis ❌ ⚠️ passports/OAuth delegation ✅ DID/VC SDK support ✅ VC issuance/verification claimed; SDK not publicly verified ⚠️ capability tokens/artifact permissions
Trust / reputation / audit ✅ capability credentials ✅ mTLS + OAuth + DID signatures + payment receipts ✅ signed receipts + reputation + policy verdicts ✅ in-path MCP enforcement + revocation + audit ✅ cryptographic ship identity + event log ⚠️ auth/protocol emphasis ⚠️ transport provenance opportunity ✅ audit + policy framing ✅ HCS audit logs + enterprise compliance positioning ✅ trust scoring, policy gates, action logging claimed ✅ handoff checkpoints + acceptance receipts
Messaging / transport ✅ Dmail ✅ A2A JSON-RPC + inbox + gateway ⚠️ gateway-oriented ⚠️ MCP proxy / control-plane oriented ✅ Ames P2P network ✅ specified + SDK ✅ email/SMS/phone ❌ / commerce and auth flow support ✅ HCS messaging + MCP/x402 agent rails ⚠️ Soulogram/verifiable messaging claimed ✅ relay + encrypted messages
Recent push 2026-07-10 2026-07-06 2026-07-10 2026-07-11 / 2026-07-05 2026-07-10 / 2026-07-10 2026-07-08 / 2026-07-05 2026-06-21 2026-07-11 / 2026-07-08 2025-01-14 spec / 2026-07-09 agent kit N/A / public SDK repos not found 2026-07-02
Stars 5 7488 28 0 / 1 3616 / 79 1347 / 326 166 30 / 2 28 spec / 36 Java SDK / 64 agent kit N/A 9

Strategic Analysis

Market Signals

  • Bindu widened its lead again. GitHub API showed Bindu at 7488★ on 2026-07-11, up from 7234★ on 2026-07-05; it remains the clearest packaged agent identity / A2A / auth / payments pressure.
  • Authority narrowing and receipts stayed active. Agent Passport System reached 28★ with a 2026-07-10 push; its BYO identity, monotonic delegation, gateway enforcement, commerce preflight, reputation, and signed-receipt language remains the sharpest benchmark.
  • MCP enforcement and enterprise governance are now explicit competitive surfaces. Chancery and AgentValet are tiny, but they name the buyer language: IdP, IGA, SPIFFE, AuthZEN, CIBA, instant revocation, MCP proxy/enforcement, and tamper-evident audit.
  • ANP remains active, but AgentConnect’s star count slipped slightly. The ANP spec repo reached 1347★ and pushed 2026-07-08; AgentConnect sits at 326★ with a 2026-07-05 push, keeping DID-WBA implementation pressure alive.
  • Transport keeps compounding. AgenticMail reached 166★ and still owns the practical email/SMS/phone-call rail story.
  • Sovereign compute is still a serious adjacent category. Urbit pushed on 2026-07-10; Kestrel also pushed on 2026-07-11 with portable DID identity, memory, and constitutional governance framing.
  • Enterprise auth/compliance language is getting sharper. Airlock, Chancery, AgentValet, Grantex, Attestix, Digital Bazaar’s agent credential server, and APS all converge on scope, revocation, audit, MCP enforcement, OAuth/OIDC bridges, and compliance.
  • Hedera remains more active in agent tooling than in the old DID repos. did-method and did-sdk-java are stable/quiet, while hedera-agent-kit-js pushed 2026-07-09 and sits at 64★.
  • Soulverse adds institutional pre-execution validation pressure, but not a verified decentralized DID root. Its live site names agentic validation, capability envelopes, model attestations, credential-gated execution, and settlement requirements, but public SDK/package availability was not verified on npm/GitHub. Cypher’s live-presentation report says the team described decentralization as a future consideration.
  • The strategic battleground is no longer DID creation. It is who owns the root authority, who enforces delegated action at tool/API boundaries, who produces receipts, and which transport/payment/audit rails developers actually integrate.

Archon Differentiators

  1. Content-addressed DID method (did:cid) — stronger integrity story than purely generative identifiers.
  2. Decentralized registry (Hyperswarm) — no central relay or hosted service as the default trust anchor.
  3. Identity + credential stack — not just identifier generation, but issuance and verification architecture.
  4. Substrate independence — identity continuity across host/model transitions remains a uniquely important framing.
  5. Multiparty governance model — separation between registry integrity and key custody is still a strong architectural story.
  6. Axionic framing — Archon has a deeper philosophical basis for agent sovereignty than most adjacent projects.
  7. Potential Lightning/payment adjacency — the emerging commerce/authorization projects make payment-aware identity materially more important.
  8. Substrate optionality — Archon can use DID authority across Lightning, HCS, x402, A2A, and other settlement/audit/transport rails instead of binding identity to one hosted platform or ledger.

Opportunities

  • Reframe Archon as the sovereign root of authority for agent actions. Identity should be explained as the anchor for scoped delegation, receipts, messaging, and payments.
  • Write a Bindu collaboration / bridge note. Explain where did:cid can complement did:bindu as the sovereign root while preserving Bindu-style A2A/x402/inbox DX.
  • Write an Urbit compatibility note. Explain how did:cid authority, credentials, Dmail, Lightning settlement, and Archon receipts could map onto Urbit-hosted services without requiring Archon to become an Urbit app first.
  • Write an ANP / AgentConnect compatibility note. Explain where did:cid and did:wba differ, where they can bridge, and why decentralized identity substrate should remain separable from communication protocol umbrellas.
  • Use AgenticMail as the transport integration story. Show DID-backed provenance over email/SMS/voice rails.
  • Use APS, Chancery, AgentValet, Soulverse, Grantex, Airlock, Credat, and HelixID as authorization/enforcement benchmarks. Archon needs equally legible examples for “this agent may do X because Y granted Z,” plus gateway enforcement and signed receipt examples.
  • Develop a receipt narrative. APS, Motebit, IDProva, Digital Bazaar, and AgentNexus make signed action receipts intuitive; Archon should connect credentials/capabilities to verifiable action logs and paid work receipts.
  • Track P2P discovery and collaboration layers. A2AL, Chorus, AgentNexus, and Urbit may become integration surfaces or proof that decentralized discovery/workflow is re-entering the conversation.
  • Write a did:cid vs did:hedera comparison. The useful contrast is not “DID vs no DID”; it is content-addressed sovereign identity and substrate independence versus Hedera/HCS-anchored identity plus enterprise-grade settlement/audit infrastructure.
  • Consider Hedera as an optional audit/payment rail. HCS and x402 could be integration targets for Archon receipts or paid API flows without making Hedera the root of agent identity.

Threats

  • Authorization-first language may outcompete identity-first language. Developers may choose the tool that answers “can this agent do this, will the gateway enforce it, and is there a signed receipt?” before caring how the DID is constructed.
  • Bindu can absorb the whole DX narrative. It offers the one-call agent-server story Archon still needs to make equally concrete.
  • Urbit can absorb sovereign-compute mindshare. For developers who already accept the personal-server/network premise, Urbit may look like the place agents should live, making Archon seem like a narrower identity/credential layer unless the bridge story is explicit.
  • ANP + AgentConnect can absorb protocol mindshare. A broad ecosystem with SDKs can become the default even if its identity substrate is less sovereign.
  • Service-oriented, IdP, IGA, trust-protocol, or gateway-centered authorization layers may move faster than decentralized infrastructure. APS, Chancery, AgentValet, Soulverse, Grantex, and Airlock-style models provide clearer DX and commercial enforcement stories; Soulverse is especially relevant because its current did:soul root appears controlled while decentralization is only future-considered.
  • Transport-layer products can become de facto identity systems. AgenticMail-style addresses and phone numbers may become practical identities unless DID-backed provenance is easy.
  • The field is fragmenting quickly. did:bindu, did:wba, did:cid, did:hedera, did:cdi, did:aip, Ed25519 receipt systems, scoped credentials, AIDs, passports, HCS/x402 rails, and ZK humanity proofs all compete for mental space.

Action Items

Immediate

  • Update Archon’s public explanation from “DID stack” to sovereign identity and authority substrate for agent action.
  • Publish a short comparison: identity substrate vs authorization layer vs communication protocol vs transport rail.
  • Add a direct Bindu collaboration note covering did:bindu, Hydra-centered DID metadata, A2A, x402, inbox UX, and where Archon should bridge versus compete.
  • Add a direct Urbit response covering Urbit ID/Azimuth, Ames, personal-server hosting, and where Archon should bridge versus compete.
  • Add a direct ANP / AgentConnect response covering did:wba, interoperability, and where Archon should complement rather than duplicate.
  • Add a direct did:cid / did:hedera comparison covering DID method semantics, registry substrate, governance, service endpoints, VC support, and payment/audit integrations.
  • Draft an authorization example using Archon credentials: user → agent → delegated action → verifiable receipt.
  • Keep Bindu, Agent Passport System, Chancery, AgentValet, Soulverse, AgentNexus, Kestrel, Airlock, AgenticMail, Hedera, Grantex, Motebit, Credat, HelixID, IDProva, A2AL, and Chorus on the watchlist.

Partnership / integration

  • Explore Bindu/A2A and APS-style gateway enforcement as first bridge targets: Archon identities should be able to sign, authorize, narrow delegation, settle, and produce receipts inside external agent runtimes without giving up did:cid as root authority.
  • Explore Urbit as a sovereign-compute host or bridge target for Archon identity, Dmail, Lightning settlement, and verifiable receipts.
  • Explore AgenticMail as a transport layer for DID-backed communications.
  • Explore Attestix as the compliance counterpart to Archon identity.
  • Evaluate whether APS/Chancery/AgentValet/Grantex/Airlock/Credat/HelixID patterns can be mapped cleanly to Archon capabilities and MCP/tool enforcement.
  • Evaluate Hedera HCS/x402 as optional settlement/audit rails for Archon credentials and receipts.
  • Evaluate Soulverse as an institutional VC/trust-protocol bridge candidate, while keeping Archon differentiated on did:cid and Lightning-native settlement.
  • Study A2AL/Chorus for decentralized discovery/messaging alignment.

Product / DX

  • Tighten the one-sentence public explanation of Archon.
  • Ship a small demo that proves delegated authority, not just DID creation.
  • Add public material explaining why decentralized registry design matters when agents act across hosts, tools, and payment rails.

Discovery Log

Date Project Source Notes
2026-02-14 payelink-agent-identity-sdk GitHub search Part of PayeLink agent suite
2026-02-14 agent-did GitHub search Closest W3C-compliant competitor
2026-02-14 agent-identity-hub GitHub search Platform/orchestration layer
2026-02-24 clawdentity GitHub API Custom did:cdi, cross-platform messaging, formal RFC spec
2026-02-24 Attestix GitHub API EU AI Act compliance, MCP-heavy positioning
2026-02-24 AgenticMail GitHub API Email/SMS infrastructure
2026-02-24 AIP GitHub API Trust chains + encrypted messaging
2026-02-24 didit-agent-skills GitHub API KYC wrappers, later resolves to didit-protocol/skills
2026-04-06 Agent Network Protocol (ANP) GitHub repo review High-traction protocol/spec project centered on did:wba and open agent networking
2026-04-06 full metadata refresh GitHub API + README check Updated stars, repo status, recent pushes, branch metadata
2026-04-06 agent-identity-hub status GitHub API Repo returns 404
2026-05-23 AgentConnect GitHub API + README ANP SDK implementation path with DID-WBA authentication guide
2026-05-23 Grantex GitHub API + README Delegated authorization / commerce-passport / audit layer
2026-05-23 Motebit GitHub API + README Sovereign agent runtime with Ed25519 identity and signed receipts
2026-05-23 Credat GitHub API + README Identity, delegation, and mutual verification SDK
2026-05-23 HelixID GitHub API + README DID/VC identity and authorization layer for agents
2026-05-23 IDProva GitHub API + README Enterprise-oriented identity, delegated authority, and audit receipts
2026-05-23 A2AL GitHub API + README Decentralized agent discovery/networking with cryptographic AIDs
2026-05-23 Chorus GitHub API + README P2P encrypted communication layer with decentralized identity
2026-05-23 full metadata refresh GitHub API Updated stars, repo status, recent pushes, and positioning
2026-06-15 full metadata refresh GitHub API Updated stars, recent pushes, and positioning across tracked repos
2026-06-17 Hedera / did:hedera GitHub API + Hedera docs Added as both a direct DID-method competitor and strategic adjacent agent/payment/audit substrate
2026-06-18 Bindu GitHub API + README/docs + PyPI Added as highest-traction agent identity/communication/payment platform; did:bindu, mTLS, Hydra OAuth, A2A, x402, inbox, gateway
2026-06-20 Bindu collaboration framing GitHub API + docs review Reframed Bindu as both top DX/platform competitor and likely bridge partner: Bindu app rails + Archon did:cid root authority
2026-06-26 Urbit GitHub API + Urbit docs Added as high-traction protocol/substrate incumbent: Urbit ID/Azimuth, personal server OS, Ames P2P networking, and sovereign-compute mindshare
2026-07-05 full metadata refresh GitHub API Updated stars, pushed dates, and availability across tracked repos; Bindu 7234★, ANP 1341★, AgentConnect 327★, AgenticMail 163★, Urbit 3615/79★
2026-07-05 Agent Passport System GitHub API + README Added as authority-layer benchmark: BYO identity, monotonic delegation narrowing, gateway enforcement, commerce preflight, reputation, and signed receipts
2026-07-05 AgentNexus GitHub API + README Added as agent collaboration/workflow substrate: DID, relay, encrypted messages, vault, playbooks, artifact handoff, objective loops, acceptance receipts
2026-07-05 Kestrel Sovereign GitHub API + README Added as adjacent sovereign-agent framework: portable DID identity, persistent memory, constitutional governance
2026-07-05 Agentic Airlock GitHub API + README Added as OAuth/compliance watchlist: Ed25519/OAuth identity verification, trust score tokens, token exchange, tamper-evident audit
2026-07-11 full metadata refresh GitHub API Updated stars, pushed dates, and availability across tracked repos; Bindu 7488★, ANP 1347★, APS 28★, Grantex 30★, AgenticMail 166★, Archon pushed 2026-07-10
2026-07-11 Chancery GitHub API Promoted to tracked watchlist: agent IdP, registry, scoped delegation, in-path MCP enforcement, instant revocation, tamper-evident audit
2026-07-11 AgentValet GitHub API Promoted to tracked watchlist: IGA for AI agents, IETF AIMS, SPIFFE, AuthZEN, CIBA, credential governance, MCP proxy
2026-07-15 Soulverse Live site + Indicio announcement + npm/GitHub availability checks Added as agent-governance / pre-execution validation watchlist: W3C DID/VC claims, Trust Protocol, Soul AI Agent SDK, Indicio Network Partner signal, public SDK packages not verified
2026-07-21 MolTrust Live API + DID method spec + pricing + publications Added as most mature centralized commercial agent trust service: did:moltrust + did:web + ERC-8004, Base L2 anchoring, AAE Internet-Draft, EU AI Act packaging
2026-08-02 full metadata refresh GitHub API Updated stars/pushed across all tracked repos; Bindu 8038★, ANP 1378★, Chancery 0→25★, APS 35★ + SDK ecosystem, AgenticMail 182★
2026-08-02 Agent Name Service (ANS) GitHub API + README + IETF Datatracker Added as registry/transparency-log entrant: IETF draft-narajala-ans-00, Merkle transparency log, SCITT COSE receipts, companion trust index
2026-08-09 full metadata refresh GitHub API Updated stars/pushed across all tracked repos; Bindu 8280★, ANP 1386★, APS 39★, AgenticMail 188★, didit skills 23★; aps-conformance-suite moved to Agent-Authority-Conformance org, aps-web 404
2026-08-09 decern GitHub search + API + README Added as authorization-kernel benchmark: Rust deterministic PDP, AuthZEN API, cvc5 SMT-proven invariants, signed hash-chained offline-verifiable decision ledger, 12★ in first week
2026-08-09 MolTrust / Soulverse re-checks Live API + npm registry MolTrust still v2.5 healthy, Lightning still roadmap; Soulverse npm SDKs still 404; both IETF drafts still -00
2026-08-16 full metadata refresh GitHub API Updated stars/pushed across all tracked repos; Bindu 8525★, ANP 1390★, APS 40★, AgenticMail 194★, ANS 35★, didit 24★; Chancery flat at 25★ (second stalled cycle); decern flat at 12★
2026-08-16 Agent-Safe Pipeline GitHub search + API + README Added as authorization-boundary benchmark: independent ALLOW/ESCALATE/BLOCK policy verdict, immutable intent capture, verified human approval, single-use intent-bound grants, conformance vectors, MCP tool-gate; 469★ / 59 forks in three days
2026-08-16 APS Rust verifier + IETF drafts GitHub API + README + IETF Datatracker aeoess/agent-passport-rust verification-only crate created 2026-08-15; APS cites draft-pidlisnyi-aps-03; MolTrust AAE draft advanced to -01 (2026-08-11); draft-narajala-ans still -00
2026-08-16 MolTrust / Soulverse re-checks Live API + npm registry MolTrust still v2.5 healthy, Lightning still roadmap, meta now leads with “The trust layer for the agent economy”; repo org now MoltyCel; Soulverse npm SDKs still 404
2026-08-23 full metadata refresh GitHub API Updated stars/pushed across all tracked repos; Bindu 8685★, ANP 1402★, AgentConnect 342★, Agent-Safe Pipeline 533★, AgenticMail 204★, APS 41★, didit 26★, Kestrel 8★; Chancery flat at 25★ (third stalled cycle); decern flat at 12★ (third week); HelixID repo moved to helixid/helixid
2026-08-23 LNSAT GitHub search + API + README Signal-only: hypler-dev/LNSAT (1★, created 2026-08-20, pre-release) — open-source execution-authorization + evidence layer (intent → policy → approval → authorization → execution → receipt → audit); structurally similar to Agent-Safe Pipeline, below profile bar
2026-08-23 MolTrust / Soulverse re-checks Live API + npm registry MolTrust API still v2.5 healthy, did:web resolves, meta unchanged; /pricing returns 403 to curl (UA blocking unconfirmed, Lightning status unverified); IETF drafts unchanged (AAE -01, ANS -00, APS -03); Soulverse npm SDKs still 404
2026-08-30 full metadata refresh GitHub API Updated stars/pushed across all tracked repos; Bindu 8685→9347★ (+662, crossed 9,000); Agent-Safe Pipeline 533→530★ (first decline); ANP 1407★; AgenticMail 212★; APS 42★; ANS 37/30★; decern 12→13★ (off flat); HelixID 4★; Chancery flat at 25★ (fourth stalled cycle); Grantex/Kestrel/APS-conformance pushed today
2026-08-30 Trusgent/AgentID GitHub search + API + README Signal-only: new open did:agent method (spec v0.1, Ed25519, blockchain-free v0.1, pip package, Apache-2.0), 1★ created 2026-08-28 — another community DID-method entry in the tail
2026-08-30 clared-ai/clared GitHub search + API + README Signal-only: bounded execution sessions — run-bound signed delegation, cumulative typed budgets, staged settlement adapters, seal-time revalidation, Ed25519-signed outcome evidence; 1★ created 2026-08-28; third intent→authority→evidence execution-boundary entrant in three weeks (after Agent-Safe Pipeline, LNSAT)
2026-08-30 MolTrust / Soulverse re-checks Live API + npm registry MolTrust API v2.5 healthy, did:web resolves; /pricing 301→/pricing.html (200): $19–$299/mo re-verified, Lightning still roadmap; Soulverse npm SDKs 404 (sixth week); IETF drafts unchanged (AAE -01, ANS -00, APS -03)
2026-09-06 full metadata refresh GitHub API Updated stars/pushed across all tracked repos; Bindu 9347→9767★ (+420, closing on 10,000); Agent-Safe Pipeline 530→533★ (recovered); APS 43★ with full-stack pushes 2026-09-04/05; ANS 38/31★; AgenticMail 218★; ANP 1413★; AgentConnect 345★; HelixID 5★; Chancery flat at 25★ (fifth stalled cycle — demoted to inactive); decern flat at 13★; Urbit 3617/81★
2026-09-06 mocenslabs/legatio-ai GitHub search + API + README Signal-only: “Legatio AI” (1★, created 2026-08-19) — Django authorization/policy/audit gateway for agents acting on behalf of humans (deterministic rules, human approval, audit trails, phase 0); fourth authorization-boundary-flavored entrant in the recent cluster (after Agent-Safe Pipeline, LNSAT, clared)
2026-09-06 microsoft/identity-spiffe GitHub search + API Signal-only: 11★, created 2026-05-26, pushed 2026-09-01 — Microsoft OSS sidecar-enforced agent-to-agent authorization bridging Entra Agent Identity and SPIFFE/SPIRE with cross-cloud workload federation; first incumbent-OSS entry in the agent-authorization space
2026-09-06 MolTrust / Soulverse re-checks Live API + npm registry MolTrust API v2.5 healthy, did:web resolves; /pricing now 301→/pricing/→403 (even browser UA); /pricing.html still 200: $19–$299/mo re-verified, Lightning still roadmap; Soulverse npm SDKs 404 (seventh week); IETF drafts unchanged (AAE -01, ANS -00, APS -03)
2026-09-13 full metadata refresh GitHub API Updated stars/pushed across all tracked repos; Bindu 9767→9829★ (+62 — cooled from +420); Agent-Safe Pipeline 533→532★ (second decline in three weeks, pushed today); APS 45★ (pushed 2026-09-10); ANP 1426★ / AgentConnect 347★ (both pushed 2026-09-12); ANS 40/31★; AgenticMail 219★ (pushed today); Chancery flat at 25★ (sixth stalled cycle); decern flat at 13★ (no push since 2026-08-24); did-sdk-java 37★; Urbit 3617/81★
2026-09-13 MolTrust AAE draft advance IETF Datatracker draft-kroehl-agentic-trust-aae advanced -01 → -02 (Datatracker dated 2026-09-06) — the only standards-track movement in the set this cycle; ANS still -00, APS still -03
2026-09-13 MolTrust / Soulverse re-checks Live API + npm registry MolTrust API v2.5 healthy, did:web resolves, meta unchanged; /pricing still 301→/pricing/→403; /pricing.html still 200: $19–$299/mo re-verified, Lightning still roadmap; Soulverse npm SDKs 404 (eighth week)
2026-09-13 discovery sweep GitHub search technocore/$FLOP airdrop-farming cluster still dominates the new-repo tail (4 repos this week — noise); two 0★ signals logged only: Stanglovicc/anlora-arp (cryptographically-signed brand claims protocol, 2026-05-14) and amulyavarshney/AgentGuard (runtime security/governance: authorization, policy enforcement, 2026-07-18); no new entrant above the profile bar
2026-09-27 full metadata refresh GitHub API Updated stars/pushed across all tracked repos (no refresh ran 2026-09-20); Bindu 9829→9809★ (first observed weekly decline, -20, forks 448); Agent-Safe Pipeline 532→589★ (+57, sharpest weekly gain since launch week, pushed 2026-09-26); ANP 1435★/AgentConnect 351★; AgenticMail 229★; ANS 43/32★; Grantex 34★/Attestix 18★ (both pushed today); Motebit 7★/HelixID 8★ (both pushed today); Chancery 25→26★ with no push (seventh stalled-push cycle); AgentValet 1→2★ (first push since 2026-08-12); decern dipped 13→12★ (no push since 2026-08-24, fourth stalled week); Urbit 3619/80★
2026-09-27 SIQ Agent Security GitHub search + API + README Added as new authorization-boundary watchlist entrant: maoyadongsh/siq-agent-security, 51★/12 forks, Apache-2.0, created 2026-08-13 — local-first agent/skill authorization runtime with Ed25519-signed intent/parameter/receipt binding, Skill Execution Context permission boundary, and an enterprise control-plane variant; highest-traction discovery since Agent-Safe Pipeline
2026-09-27 MolTrust / Soulverse re-checks Live API + npm registry MolTrust API v2.5 healthy, did:web resolves, meta unchanged; /pricing still 301→/pricing/→403; /pricing.html still 200: $19–$299/mo re-verified, Lightning still roadmap; Soulverse npm SDKs 404 (ninth week); IETF drafts unchanged (AAE -02, ANS -00, APS -03)
2026-10-04 full metadata refresh GitHub API Updated stars/pushed across all tracked repos; Bindu 9809→10073★ (+264, crossed 10,000 — but no push since 2026-09-06; forks 448→635); Attestix 18→875★ (anomalous +857 spike, unverified traction); ANP 1435→1441★ / AgentConnect 351★; AgenticMail 229→231★; Agent-Safe Pipeline flat at 589★ (pushed today); SIQ flat at 51★; APS flat at 46★ (pushed today); Grantex 34★ / Kestrel 8★ / Motebit 7★ (all pushed today); decern flat at 12★ (fifth stalled week); Chancery flat at 26★, no push (eighth stalled cycle); ANS 43/32★; didit 26→27★; Hedera Agent Kit 68→69★; Urbit 3619→3620★ / vere 80★; Archon 5→6★
2026-10-04 APS org move + draft advance GitHub API + IETF Datatracker aeoess/agent-passport-system and aeoess/agent-passport-rust now resolve under a dedicated agent-passport-system org; draft-pidlisnyi-aps advanced -03 → -04 (Datatracker dated 2026-09-28); AAE still -02, ANS still -00
2026-10-04 discovery sweep GitHub search No new entrant above the profile bar; signal-only: ndrorchestration/DGAF-Framework (4★, evidence-bound AI governance framework), opena2a-standards/agent-authorization-protocol (1★, scoped/attested authorization draft), frostyjay7813/AgentFence (0★, authorization-bound execution); technocore/$FLOP cluster persists (noise); microsoft/identity-spiffe pushed 2026-10-03 (11★)
2026-10-04 MolTrust / Soulverse re-checks Live API + npm registry MolTrust API still v2.5 healthy (2026-10-04 13:03 UTC), did:web resolves, homepage meta unchanged; /pricing still 301→/pricing/→403; /pricing.html still 200: $19–$299/mo re-verified, Lightning still roadmap; Soulverse npm SDKs 404 (tenth consecutive weekly check)
2026-10-11 full metadata refresh GitHub API Updated stars/pushed across all tracked repos; Agent-Safe Pipeline 589→687★ (+98, sharpest move — and repo description repositioned to an installable execution-authority gateway: Homebrew/Linux/Docker/Helm); Bindu 10073→10076★ (+3, growth collapsed; no push since 2026-09-06, fifth week; forks 635→669); decern 12→13★ (broke five-week push stall, pushed 2026-10-10); Attestix 875→874★ (spike not compounding, no push since 2026-10-01); APS 46→49★ (pushed 2026-10-10); ANP 1441→1444★ / AgentConnect 351→352★ (pushed 2026-10-09 / today); AgenticMail 231→236★; SIQ flat at 51★ (pushed 2026-10-09); Grantex 34→33★; ANS 43/32★ (pushed 2026-10-08 / 2026-10-10); Chancery flat at 26★, no push (ninth stalled cycle); didit 27→28★; Hedera Agent Kit 69★ (pushed 2026-10-07); Urbit 3620→3622★ / vere 80★; MoltyCel/moltrust-api 3→4★ (pushed today); Motebit/Kestrel pushed today; Archon flat at 6★
2026-10-11 microsoft/identity-spiffe gone GitHub API Signal-only item now returns 404 (was 11★, last observed pushed 2026-10-03); the first incumbent-OSS agent-authorization entry is no longer publicly visible — repo removed, renamed, or made private
2026-10-11 discovery sweep GitHub search No new entrant above the profile bar (all ≤2★). Signal-only: wienerlabs/square-solana (2★, compliance-gated agent job settlement on Solana — USDC escrow, ZK spending-mandate proofs, DID identity), oleg-tkachuk/limes (0★, Biscuit budgeted/delegable capabilities, Go), thesithunyein/witness (0★, passkey-signed on-chain human authorization, Monad/ERC-8004), DuNorth595/xrpl_agent_id (1★, did:xrpl + XLS-70 VCs on XRPL), goun7/Kredent (0★, did:key + reputation)
2026-10-11 MolTrust / Soulverse re-checks Live API + npm registry MolTrust API v2.5 healthy (2026-10-11 13:03 UTC), did:web resolves, meta unchanged; /pricing still 301→/pricing/→403; /pricing.html still 200: $19–$299/mo re-verified, Lightning still roadmap — third quiet live-surface cycle; Soulverse npm SDKs 404 (eleventh week); IETF drafts unchanged (AAE -02, APS -04, ANS -00)
2026-10-15 full metadata refresh GitHub API Updated stars/pushed across all tracked repos; Bindu 10076→10058★ (-18 — first star decline since tracking began; no push since 2026-09-06, sixth week; forks 669→670); Agent-Safe Pipeline 687→688★ (+1 — spike flattened, pushed 2026-10-13); AgentConnect flat at 352★ but primary language shifted Python→Rust (pushed 2026-10-14); ANP 1444→1450★ (pushed 2026-10-15); Attestix flat at 874★ (resumed pushes 2026-10-15); SIQ 51→50★; AIP 15→14★; Hedera did-method 28→27★ / did-sdk-java 37→36★; ANS 43/33★ (both pushed 2026-10-15); AgenticMail 236→237★; APS/Grantex/Motebit/Kestrel/Urbit (3622/80) all pushed 2026-10-14/15; Chancery flat at 26★, no push (tenth stalled cycle); Archon flat at 6★ (pushed 2026-10-14)
2026-10-15 discovery sweep GitHub search No new entrant above the profile bar (all 0★). Signal-only: harshittDEV/ScopeLock-AI (0★, agent authorization firewall), gatekeeper-os/private-auth (0★, request-bound agent authorization with local Semaphore proofs + Ethereum permission registry), m7mdd77/threshold-decision-engine (0★, evidence-driven authorization with execution-time rechecks and sandbox receipts); the authorization-boundary cluster continues to spawn zero-traction entrants
2026-10-15 MolTrust / Soulverse re-checks Live API + npm registry MolTrust API v2.5 healthy (2026-10-15 17:24 UTC), did:web resolves, meta unchanged; /pricing still 301→/pricing/→403; /pricing.html still 200: $19–$299/mo re-verified, Lightning still roadmap — fourth quiet live-surface cycle; Soulverse npm SDKs 404 (twelfth week); IETF drafts unchanged (AAE -02, APS -04, ANS -00)

Research Questions

  1. Which teams need a sovereign identity substrate versus an authorization layer, communication protocol, transport rail, trust layer, or compliance wrapper?
  2. Where does Archon most clearly outperform alternatives in practice: decentralization, governance, credentials, migration resilience, receipts, or payment-aware authority?
  3. What public comparison best explains why identity, authorization, messaging, and transport should be separated but composable?
  4. Should Archon bridge to did:wba / AgentConnect, compete with it, or remain explicitly substrate-focused?
  5. How should Archon position did:cid against did:hedera: sovereign root of authority, chain-independent registry, credential/service-endpoint model, or payment/audit composability?
  6. How should Archon position against Urbit: bridge to Urbit-hosted services, compete as a lighter agent-authority layer, or treat Urbit as a separate sovereign-compute ecosystem?
  7. Which adjacent project is the best first integration story: Bindu/A2A, APS gateway enforcement, Urbit, AgenticMail, Attestix, Hedera HCS/x402, Grantex/Credat/HelixID, Airlock/OAuth, A2AL/Chorus/AgentNexus, or ANP/AgentConnect?

Additional Resources

This is a living document. Refresh repo metadata weekly, revisit feature claims monthly, and log availability changes immediately when tracked repos disappear, redirect, or materially reposition themselves.