Update compliance docs: Article 86 explain(), on_event_async, monitoring

Aligns with flowscript-agents 0.2.7 release:
- Enterprise & Compliance table updated (Art 86 → explain(), Art 72 → on_event)
- Learn page: audit section expanded with Article 86 and SIEM async
- TS README notes Python explain() with v1.2 TS port planned

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: phillipclapham

README.md

@@ -317,13 +317,16 @@ Vector search and reasoning queries are orthogonal — use both. Mem0 for retrie
 
 FlowScript's typed reasoning chains are also compliance-ready audit infrastructure. This isn't a separate product — it's a structural property of how FlowScript works.
 
-**EU AI Act (Articles 12, 13, 86):**
+**EU AI Act coverage:**
 
 | Requirement | Article | How FlowScript satisfies it |
 |:---|:---|:---|
 | Record-keeping | Art. 12 | Hash-chained audit trail, append-only, tamper-evident, 7yr default retention |
 | Transparency | Art. 13 | `why()` queries return typed causal chains — not reconstructions, actual reasoning records |
-| Right to explanation | Art. 86 | `alternatives()` reconstructs what was considered, what was chosen, and the rationale |
+| Right to explanation | Art. 86 | `explain()` generates deterministic compliance documents (Python SDK) |
+| Monitoring | Art. 72 | `onEvent` / `on_event_async` callbacks stream audit events to SIEM/monitoring |
+
+**Article 86 — Right to Explanation:** The [Python SDK](https://pypi.org/project/flowscript-agents/) provides `explain()` — a deterministic, no-LLM function that converts `why()` results into formal compliance documents with Article 86 citation, hash-chain reference, and three audience modes (general/legal/technical). Also available as the `explain_decision` MCP tool. TypeScript port planned for v1.2.
 
 **Enforcement begins August 2026.** Audit trails can't be backdated. Organizations using FlowScript today have unbroken reasoning records from day one. You can turn on logging tomorrow — you can't manufacture the last 18 months of decision provenance.
 
@@ -337,6 +340,7 @@ FlowScript's typed reasoning chains are also compliance-ready audit infrastructu
 │  ┌──────────┐ ┌──────────┐ ┌──────────────────────┐ │
 │  │  Memory   │ │ Queries  │ │ Audit Trail          │ │
 │  │  (graph)  │ │ (5 ops)  │ │ (SHA-256 hash chain) │ │
+│  │         explain()      │ │  on_event_async      │ │
 │  └──────────┘ └──────────┘ └──────────────────────┘ │
 ├─────────────────────────────────────────────────────┤
 │  Your Storage (files, database, cloud)              │

web/src/pages/Learn.tsx

@@ -273,9 +273,9 @@ mem.query.alternatives(dbQuestion);{"\n"}
         </p>
       </section>
 
-      {/* Audit Trail */}
+      {/* Audit Trail & Compliance */}
       <section className="learn-section">
-        <h2>Audit trail</h2>
+        <h2>Audit trail &amp; compliance</h2>
         <p>
           Every mutation to the reasoning graph is recorded in a SHA-256 hash-chained, append-only audit trail.
           Each entry links to the previous via cryptographic hash. Tampering with history breaks the chain.
@@ -286,11 +286,25 @@ mem.query.alternatives(dbQuestion);{"\n"}
           <li><strong>Framework attribution:</strong> Every entry tagged with which adapter (LangGraph, CrewAI, etc.) made the change.</li>
           <li><strong>Cross-language:</strong> Same canonical JSON serialization in TypeScript and Python. Chains verify across languages.</li>
           <li><strong>Configurable retention:</strong> Default 7 years (SOX). Monthly rotation with gzip compression.</li>
-          <li><strong>SIEM integration:</strong> <code>onEvent</code> callback for streaming audit events to external systems.</li>
+          <li><strong>SIEM integration:</strong> <code>onEvent</code> / <code>on_event_async</code> callbacks stream audit events to external systems. Async mode uses a background thread so slow webhooks never block agent operations.</li>
         </ul>
+
+        <h3>EU AI Act &mdash; Article 86 (Right to Explanation)</h3>
+        <p>
+          The Python SDK includes <code>explain()</code> &mdash; a deterministic, no-LLM function that converts{" "}
+          <code>why()</code> results into formal compliance documents. Three audience modes: plain English for
+          affected individuals, formal compliance language for regulatory submissions (with Article 86 citation
+          and hash-chain verification reference), and structured output for developers. Same input always produces
+          the same output. Also available as the <code>explain_decision</code> MCP tool.
+        </p>
+        <p>
+          Enforcement begins August 2026. Audit trails can't be backdated. Organizations using FlowScript today
+          have unbroken reasoning records from day one.
+        </p>
+
         <p>
-          You can't audit a deletion. You can query a tension. This is the structural argument for RELATE &gt; DELETE
-          . FlowScript preserves the full reasoning record, not just the latest version.
+          You can't audit a deletion. You can query a tension. This is the structural argument for RELATE &gt; DELETE.
+          FlowScript preserves the full reasoning record, not just the latest version.
         </p>
       </section>