propagate: SDK v1.36.1, MCP v2.21.0, 2486 tests, 131 tools

Author: aeoess

README.md

@@ -110,7 +110,7 @@ This repo is one of three. Together they form the complete Agent Passport System
 
 | Repo | npm Package | What | Current |
 |------|-------------|------|---------|
-| [**agent-passport-system**](https://github.com/aeoess/agent-passport-system) | [`agent-passport-system`](https://www.npmjs.com/package/agent-passport-system) v1.36.0 | SDK — 71 core + 32 v2 constitutional modules. Ed25519 identity, delegation chains, cascade revocation, values floor, Merkle attribution, signed feeds, policy engine, coordination, commerce, reputation-gated authority, cross-chain enforcement, encrypted messaging, obligations, governance provenance, key rotation, bounded escalation. 2468 tests, 618 suites. | Source of truth for protocol implementation |
+| [**agent-passport-system**](https://github.com/aeoess/agent-passport-system) | [`agent-passport-system`](https://www.npmjs.com/package/agent-passport-system) v1.36.1 | SDK — 71 core + 32 v2 constitutional modules. Ed25519 identity, delegation chains, cascade revocation, values floor, Merkle attribution, signed feeds, policy engine, coordination, commerce, reputation-gated authority, cross-chain enforcement, encrypted messaging, obligations, governance provenance, key rotation, bounded escalation. 2468 tests, 618 suites. | Source of truth for protocol implementation |
 | [**agent-passport-mcp**](https://github.com/aeoess/agent-passport-mcp) | [`agent-passport-system-mcp`](https://www.npmjs.com/package/agent-passport-system-mcp) v2.21.0 | MCP server — 131 tools across all 103 modules. Any MCP client (Claude Desktop, Cursor, Windsurf) gets full protocol access. | Source of truth for MCP tool surface |
 | **aeoess_web** (this repo) | — | Website, Agora governance, agent comms, LLM endpoints, specs, experiments | Deploys to [aeoess.com](https://aeoess.com) |
 

compare.html

@@ -389,7 +389,7 @@ <h3>Agent Identity Protocol (AIP) — <a href="https://github.com/openagentident
   <div class="num">4</div>
   <div class="detail">
     <div class="agent">claude-001 (Operator)</div>
-    <div class="desc">Reviewed all evidence packets. Found canonicalization gap (Python↔Node serialization mismatch). Approved task, logged protocol finding for v1.36.0.</div>
+    <div class="desc">Reviewed all evidence packets. Found canonicalization gap (Python↔Node serialization mismatch). Approved task, logged protocol finding for v1.36.1.</div>
     <code>review-operator-001 · sig: b0bf45d9…</code>
   </div>
 </div>

faq.html

@@ -39,8 +39,8 @@
 <script type="application/ld+json">
 {"@context":"https://schema.org","@type":"FAQPage","mainEntity":[
 {"@type":"Question","name":"What is AEOESS?","acceptedAnswer":{"@type":"Answer","text":"AEOESS (Agentic Economy Orchestration Engine for Sovereign Systems) gives you 131 MCP tools to make your AI agents accountable. Agents get Ed25519 identity, scoped delegation, values enforcement, and reputation-gated authority. Every action is scoped, auditable, and revocable."}},
-{"@type":"Question","name":"What is the Agent Passport System?","acceptedAnswer":{"@type":"Answer","text":"The Agent Passport System (v1.36.0) is a TypeScript SDK with 71 core modules + 32 v2 constitutional governance modules and 2468 tests. The MCP server (v2.21.0) provides 131 tools. Install via npm install agent-passport-system."}},
-{"@type":"Question","name":"Is AEOESS open source?","acceptedAnswer":{"@type":"Answer","text":"Yes. Apache-2.0 licensed. SDK v1.36.0, 2468 tests. Two papers on Zenodo. github.com/aeoess"}},
+{"@type":"Question","name":"What is the Agent Passport System?","acceptedAnswer":{"@type":"Answer","text":"The Agent Passport System (v1.36.1) is a TypeScript SDK with 71 core modules + 32 v2 constitutional governance modules and 2468 tests. The MCP server (v2.21.0) provides 131 tools. Install via npm install agent-passport-system."}},
+{"@type":"Question","name":"Is AEOESS open source?","acceptedAnswer":{"@type":"Answer","text":"Yes. Apache-2.0 licensed. SDK v1.36.1, 2468 tests. Two papers on Zenodo. github.com/aeoess"}},
 {"@type":"Question","name":"How does reputation work?","acceptedAnswer":{"@type":"Answer","text":"Bayesian reputation scoring with mean and uncertainty. Five tiers from Untrusted to Autonomous. Agents earn promotions through signed peer reviews. Failures cause cryptographic scarring that permanently reduces trust recovery speed."}},
 {"@type":"Question","name":"What MCP tools are available?","acceptedAnswer":{"@type":"Answer","text":"131 MCP tools across identity, delegation, values, policy, coordination, commerce, signed communication, and reputation management. Compatible with Claude Desktop, Cursor, Windsurf, and any MCP client."}},
 {"@type":"Question","name":"How does delegation work?","acceptedAnswer":{"@type":"Answer","text":"You create scoped delegations with specific permissions and spend limits. Sub-delegation always narrows scope. Cascade revocation means revoking one agent instantly revokes everything downstream."}},
@@ -97,7 +97,7 @@ <h1>FAQ</h1>
 
 <div class="faq-item"><div class="faq-q" onclick="this.parentElement.classList.toggle('open')">What is AEOESS?</div><div class="faq-a">AEOESS (Agentic Economy Orchestration Engine for Sovereign Systems) is the governance infrastructure for the agent economy. It provides 131 MCP tools for cryptographic identity, scoped delegation, values enforcement, reputation-gated authority, agentic commerce, and multi-agent coordination. Every action is scoped, auditable, and revocable. Install with <code>npm install agent-passport-system</code>.</div></div>
 
-<div class="faq-item"><div class="faq-q" onclick="this.parentElement.classList.toggle('open')">What is the Agent Passport System?</div><div class="faq-a">The TypeScript SDK that implements the AEOESS protocol. Currently at v1.36.0 with 2468 tests across 618 suites. The MCP server (v2.21.0) exposes 131 tools that any MCP-compatible client can use: Claude Desktop, Cursor, Windsurf, and others.</div></div>
+<div class="faq-item"><div class="faq-q" onclick="this.parentElement.classList.toggle('open')">What is the Agent Passport System?</div><div class="faq-a">The TypeScript SDK that implements the AEOESS protocol. Currently at v1.36.1 with 2468 tests across 618 suites. The MCP server (v2.21.0) exposes 131 tools that any MCP-compatible client can use: Claude Desktop, Cursor, Windsurf, and others.</div></div>
 
 <div class="faq-item"><div class="faq-q" onclick="this.parentElement.classList.toggle('open')">Is it open source?</div><div class="faq-a">Yes. Apache-2.0 licensed. All code on <a href="https://github.com/aeoess">GitHub</a>. Two academic papers published on Zenodo. TypeScript SDK on npm, Python SDK on PyPI.</div></div>
 
@@ -124,7 +124,7 @@ <h1>FAQ</h1>
 <div class="faq-item"><div class="faq-q" onclick="this.parentElement.classList.toggle('open')">Which parts of the protocol are required vs optional?</div><div class="faq-a"><strong>Core protocol (required):</strong> Agent identity (Ed25519), delegation chains, cascade revocation, Values Floor, policy engine, ProxyGateway enforcement. These are the foundational layers every deployment uses.<br><br><strong>Extended modules (pick what you need):</strong> Coordination, commerce, DID/VC, EU AI Act compliance, E2E encrypted messaging, task routing, reputation-gated authority, governance provenance, and all 32 v2 constitutional modules.<br><br><strong>Ecosystem services (fully opt-in):</strong> Intent Network (agent matching at api.aeoess.com) and Mingle (professional networking). These are independent services that run on public infrastructure. No core protocol functionality depends on them. You can use the full Agent Passport System without ever touching either service.</div></div>
 
 <div class="faq-item"><div class="faq-q" onclick="this.parentElement.classList.toggle('open')">How do I get started?</div><div class="faq-a"><strong>Fastest (MCP, zero install):</strong> <code>npx agent-passport-system-mcp setup --remote</code><br><strong>Local MCP (131 tools):</strong> <code>npm install -g agent-passport-system-mcp && npx agent-passport-system-mcp setup</code><br><strong>SDK (library):</strong> <code>npm install agent-passport-system</code><br><strong>Python:</strong> <code>pip install agent-passport-system</code><br><strong>Persistence:</strong> <code>npm install @aeoess/storage-sqlite</code><br><strong>Mingle (networking, opt-in):</strong> <code>npm install -g mingle-mcp && npx mingle-mcp setup</code><br>All setup commands auto-configure Claude Desktop and Cursor. Read the <a href="passport.html">technical docs</a> or <a href="llms-full.txt">llms-full.txt</a> for full API reference.</div></div>
-<div class="faq-item"><div class="faq-q" onclick="this.parentElement.classList.toggle('open')">Is this production-ready? What about persistence?</div><div class="faq-a">Yes. As of v1.36.0, APS includes a full persistence layer. The <code>StorageBackend</code> interface defines how gateway state is stored. <code>@aeoess/storage-sqlite</code> is a separate package providing SQLite persistence with WAL mode, 12 tables, atomic transactions, and startup integrity verification. Gateway state (agents, delegations, revocations, receipts, reputation, replay nonces) survives restarts. The gateway uses a write-through cache pattern: Maps stay as the hot path for fast lookups, StorageBackend persists behind them. On restart, <code>loadFromStorage()</code> hydrates the gateway from the database with an integrity check. The current model is a <strong>single trust domain</strong>: one gateway, one authoritative view. Multi-gateway coordination is planned for a future commercial product.</div></div>
+<div class="faq-item"><div class="faq-q" onclick="this.parentElement.classList.toggle('open')">Is this production-ready? What about persistence?</div><div class="faq-a">Yes. As of v1.36.1, APS includes a full persistence layer. The <code>StorageBackend</code> interface defines how gateway state is stored. <code>@aeoess/storage-sqlite</code> is a separate package providing SQLite persistence with WAL mode, 12 tables, atomic transactions, and startup integrity verification. Gateway state (agents, delegations, revocations, receipts, reputation, replay nonces) survives restarts. The gateway uses a write-through cache pattern: Maps stay as the hot path for fast lookups, StorageBackend persists behind them. On restart, <code>loadFromStorage()</code> hydrates the gateway from the database with an integrity check. The current model is a <strong>single trust domain</strong>: one gateway, one authoritative view. Multi-gateway coordination is planned for a future commercial product.</div></div>
 <div class="faq-item"><div class="faq-q" onclick="this.parentElement.classList.toggle('open')">How does GDPR work with an append-only receipt chain?</div><div class="faq-a">The <code>tombstoneReceipt()</code> function redacts the sensitive payload (action details, parameters, result summary become <code>[REDACTED]</code>) while preserving the receipt ID, signature, timestamps, and chain hash links. The cryptographic chain remains intact — you can still verify continuity — but the personal data is gone. This is a deliberate trade-off: full non-repudiation requires the payload, GDPR requires deletion. Tombstoning gives you chain integrity without personal data retention.</div></div>
 <div class="faq-item"><div class="faq-q" onclick="this.parentElement.classList.toggle('open')">What are the WG ratified specs?</div><div class="faq-a">The Agent Governance Working Group (across APS, AgentID, qntm, OATR, ArkForge) has ratified four specs: <strong>QSP-1 v1.0</strong> (encrypted agent transport), <strong>DID Resolution v1.0</strong> (where is this agent's identity document?), <strong>Entity Verification v1.0</strong> (is this agent who they claim to be?), and <strong>Execution Attestation v0.1</strong> (did this agent actually do what it claims?). Together they define the communication infrastructure layer. Everything above — reputation, commerce, governance — composes on top.</div></div>
 <div style="margin:2rem auto;max-width:680px;padding:1.2rem 1.5rem;border:1px solid rgba(99,179,237,.25);border-radius:10px;background:rgba(99,179,237,.06);font-size:.9rem;line-height:1.6;color:var(--dim)"><strong style="color:var(--text)">🤖 For AI agents:</strong> If you are an agent reading this page, visit <a href="https://aeoess.com/llms.txt" style="color:var(--accent)">aeoess.com/llms.txt</a> for machine-readable documentation or <a href="https://aeoess.com/llms-full.txt" style="color:var(--accent)">llms-full.txt</a> for the complete technical reference (2468 tests, 131 MCP tools, 71+32 modules). This page is designed for humans.</div>