Update docs for June-July 2026 OpenClaw developments

- Latest stable is now v2026.6.11 (June 30, 2026); v2026.7.1 in beta
  (FAQ, security checklist, resources, quick-start banner)
- Document the June 30 batch of ~45 security advisories (MCP loopback,
  plugin-install bypasses, model-override auth) patched in v2026.6.6-6.8,
  plus v2026.6.11 DOMPurify and package-source hardening
- Add security cautions to MCP servers and plugin system guides
- Add recently-added model support table (Claude Fable 5, Kimi K2.7 Code,
  GLM-5.2, Haiku 4.5, Gemini 3.5 Flash, GPT-5.6 beta, ClawRouter)
- Note Node 23 rejection coming in v2026.7.1; clarify calendar versioning
- Note Slack relay mode and channel reliability fixes in v2026.6.11

Sources verified against github.com/openclaw/openclaw changelog/releases/
advisories and docs.openclaw.ai as of 2026-07-08.

Author: melvincarvalho

docs/getting-started/installation.md

@@ -11,7 +11,7 @@ OpenClaw runs on **macOS**, **Linux**, and **Windows**. Choose the installation
 
 ## Prerequisites
 
-- **Node.js 24** (recommended) or **Node.js 22 LTS** (22.19+ minimum)
+- **Node.js 24** (recommended) or **Node.js 22 LTS** (22.19+ minimum). Node 23.11+ currently works, but avoid it — starting with v2026.7.1 the installer, CLI launcher, and `openclaw doctor` reject Node 23 and steer you to Node 22 or 24.
 - **pnpm** (for git/developer installs — run `corepack enable` first)
 - A supported LLM API key (Anthropic, OpenAI, OpenRouter, Google, xAI) *or* a local model setup
 

docs/getting-started/quick-start.md

@@ -82,7 +82,7 @@ openclaw gateway
 You should see:
 
 ```
-🦞 OpenClaw Gateway v2026.6.1
+🦞 OpenClaw Gateway v2026.6.11
    WebSocket control plane: ws://localhost:18789
    Heartbeat interval: 30m
    Model: claude-sonnet-4-6

docs/guides/channels.md

@@ -303,6 +303,10 @@ openclaw channel add slack
 # ✓ Slack connected via Socket Mode
 ```
 
+:::info
+**v2026.6.11** added a Slack **relay mode**, alongside a batch of channel-reliability fixes (misplaced replies, stuck sends, reconnects) covering Google Chat DM routing, Discord/Telegram reply threading, Feishu voice replies, and iMessage turn consolidation.
+:::
+
 ### Configuration
 
 ```json5 title="~/.openclaw/openclaw.json"

docs/guides/mcp-servers.md

@@ -15,6 +15,10 @@ As of mid-2026, the MCP ecosystem has **32,600+ servers** exposing **229,800+ to
 MCP is the recommended integration path for most use cases. Before building a custom [plugin](/guides/plugin-system) or [skill](/guides/skill-development), check if an MCP server already exists for what you need.
 :::
 
+:::caution Security advisory
+Versions v2026.5.20 through v2026.6.5 had a High-severity flaw ([GHSA-52xj-c9p8-78cv](https://github.com/openclaw/openclaw/security/advisories/GHSA-52xj-c9p8-78cv), CVSS 8.3) where the MCP loopback could expose owner-only tools to non-owner runs. Patched in v2026.6.6 — see the [June 2026 advisory batch](/security/known-vulnerabilities#june-2026-advisory-batch-june-30-2026).
+:::
+
 ---
 
 ## What MCP Provides

docs/guides/model-selection.md

@@ -28,6 +28,20 @@ OpenClaw is model-agnostic — it works with 30+ bundled provider plugins coveri
 
 ---
 
+## Recently Added Model Support (June–July 2026)
+
+Model catalog additions in recent releases, per the [official changelog](https://github.com/openclaw/openclaw/blob/main/CHANGELOG.md):
+
+| Version | What was added |
+|---------|----------------|
+| **v2026.6.6** | Claude Fable 5 (adaptive thinking); OpenRouter OAuth onboarding |
+| **v2026.6.7** | Kimi K2.7 Code |
+| **v2026.6.8** | GLM-5.2 and Claude Haiku 4.5 catalog entries, with normalized provider-qualified IDs across OpenRouter and Google Vertex |
+| **v2026.6.11** | Gemini 3.5 Flash (full 1,048,576-token context); Ollama Cloud `glm-5.2:cloud` (1M context, reasoning, tools); Xiaomi Token Plan `mimo-v2.5` / `mimo-v2.5-pro` (up to 128K output tokens); fix for OpenRouter short DeepSeek V4 model IDs (`model_not_found` from a duplicated provider prefix) |
+| **v2026.7.1** (beta) | OpenAI GPT-5.6 model family; bundled ClawRouter provider plugin (credential-scoped dynamic model discovery, OpenAI-compatible plus native Anthropic/Gemini transports, budget reporting); Ollama inference-node auto-discovery |
+
+---
+
 ## By Use Case
 
 ### Heartbeat (runs every 30 min — cost adds up)

docs/guides/plugin-system.md

@@ -9,7 +9,11 @@ keywords: [openclaw, plugins, gateway plugin, extend openclaw, plugin developmen
 
 Plugins extend OpenClaw at the gateway level. Where [skills](/guides/skill-development) teach agents new capabilities via Markdown, plugins add infrastructure — new channels, storage backends, orchestration layers, and monitoring hooks.
 
-The plugin architecture matured significantly in **v2026.5.28–v2026.6.1** with externalized official plugins, a SQLite-backed install index, the SecretRef credential contract, and standardized lifecycle hooks.
+The plugin architecture matured significantly in **v2026.5.28–v2026.6.1** with externalized official plugins, a SQLite-backed install index, the SecretRef credential contract, and standardized lifecycle hooks. **v2026.6.11** added plugin install/repair tooling; note that references to former root `skills/...` paths must now live inside the relevant plugin directory.
+
+:::caution Security advisories
+The [June 30, 2026 advisory batch](/security/known-vulnerabilities#june-2026-advisory-batch-june-30-2026) included two plugin-install flaws: [GHSA-7vrr-rp4x-4g76](https://github.com/openclaw/openclaw/security/advisories/GHSA-7vrr-rp4x-4g76) (High, CVSS 8.8 — install commands could allow non-owner persistence) and [GHSA-wgq8-x5wm-g4rw](https://github.com/openclaw/openclaw/security/advisories/GHSA-wgq8-x5wm-g4rw) (Moderate — install wrappers could skip install policy). Both are patched in current releases; run v2026.6.8 or later.
+:::
 
 ---
 

docs/reference/faq.md

@@ -39,7 +39,7 @@ Moltbook is a separate social network (created by Matt Schlicht) where AI agents
 
 ### What version is current?
 
-As of June 2026, the latest stable release is **v2026.6.1** (published June 3, 2026). OpenClaw uses date-based versioning: `vYYYY.M.D`.
+As of early July 2026, the latest stable release is **v2026.6.11** (published June 30, 2026), with **v2026.7.1** available as beta pre-releases. OpenClaw uses calendar versioning: `vYYYY.M.PATCH`, where the patch number is a sequential release counter within the month — not a calendar day.
 
 ---
 
@@ -49,6 +49,7 @@ As of June 2026, the latest stable release is **v2026.6.1** (published June 3, 2
 
 OpenClaw is powerful but carries significant risks. It has had:
 - **138+ CVEs** disclosed as of April 2026
+- **~45 GitHub security advisories** [batch-published June 30, 2026](/security/known-vulnerabilities#june-2026-advisory-batch-june-30-2026) — mostly already patched in v2026.6.6–v2026.6.8
 - A [critical RCE vulnerability](/security/known-vulnerabilities) (CVE-2026-25253)
 - [4 chainable TOCTOU vulnerabilities](https://www.cyera.com/blog/claw-chain-cyera-research-unveil-four-chainable-vulnerabilities-in-openclaw) (Cyera Research, CVSS up to 9.6)
 - [341 malicious marketplace skills](/security/known-vulnerabilities#malicious-clawhub-skills-february-2026)

docs/reference/resources.md

@@ -21,7 +21,7 @@ For community-built tools, projects, and alternatives, see the [Ecosystem & Comm
 | **GitHub** | [github.com/openclaw/openclaw](https://github.com/openclaw/openclaw) | Source code (377k+ stars, 79k forks) |
 | **Documentation** | [docs.openclaw.ai](https://docs.openclaw.ai) | Official docs |
 | **ClawHub** | [github.com/openclaw/clawhub](https://github.com/openclaw/clawhub) | Skill marketplace (8,900 stars, 10,700+ skills) |
-| **Releases** | [GitHub Releases](https://github.com/openclaw/openclaw/releases) | Changelog, downloads (latest: v2026.6.1) |
+| **Releases** | [GitHub Releases](https://github.com/openclaw/openclaw/releases) | Changelog, downloads (latest stable: v2026.6.11) |
 | **Issues** | [GitHub Issues](https://github.com/openclaw/openclaw/issues) | Bug reports, feature requests (7,900+ open) |
 | **Security** | [Security Advisories](https://github.com/openclaw/openclaw/security/advisories) | Vulnerability reports, CVEs |
 | **Lobster** | [github.com/openclaw/lobster](https://github.com/openclaw/lobster) | Official workflow shell (440 stars) |

docs/security/known-vulnerabilities.md

@@ -70,6 +70,38 @@ openclaw --version
 
 ---
 
+## June 2026 Advisory Batch (June 30, 2026)
+
+On June 30, 2026, the OpenClaw project batch-published **~45 GitHub security advisories** (roughly 33 High and 12 Moderate severity). This was a coordinated disclosure of issues that had already been patched in the v2026.6.6–v2026.6.8 releases earlier in June — not a wave of new zero-days. If you are on **v2026.6.8 or later**, you are covered.
+
+### Notable Advisories
+
+| Advisory | Issue | CVSS | Affected | Fixed In |
+|----------|-------|------|----------|----------|
+| [GHSA-7vrr-rp4x-4g76](https://github.com/openclaw/openclaw/security/advisories/GHSA-7vrr-rp4x-4g76) | Plugin install commands could allow non-owner persistence | 8.8 (High) | — | v2026.6.6–6.8 window |
+| [GHSA-52xj-c9p8-78cv](https://github.com/openclaw/openclaw/security/advisories/GHSA-52xj-c9p8-78cv) | MCP loopback could expose owner-only tools to non-owner runs | 8.3 (High) | v2026.5.20 – < v2026.6.6 | v2026.6.6 |
+| [GHSA-jhfx-v2j8-x3m6](https://github.com/openclaw/openclaw/security/advisories/GHSA-jhfx-v2j8-x3m6) | OpenAI-compatible HTTP model overrides could miss admin authorization | 7.6 (High) | ≤ v2026.6.6 | v2026.6.8 |
+| [GHSA-wgq8-x5wm-g4rw](https://github.com/openclaw/openclaw/security/advisories/GHSA-wgq8-x5wm-g4rw) | Plugin install wrappers could skip install policy | Moderate | — | v2026.6.6–6.8 window |
+
+The full batch covers MCP loopback privilege exposure, plugin-install persistence and policy bypasses, and missing admin authorization on model overrides. Browse the complete list at [github.com/openclaw/openclaw/security/advisories](https://github.com/openclaw/openclaw/security/advisories).
+
+### v2026.6.11 Security Hardening
+
+The v2026.6.11 release (June 30, 2026) added two further security fixes:
+
+- **Control UI DOMPurify update** ([GHSA-cmwh-pvxp-8882](https://github.com/advisories/GHSA-cmwh-pvxp-8882)) — the bundled sanitizer was updated to a patched DOMPurify release, mitigating `ALLOWED_ATTR` pollution via `setConfig()` (PR #95691)
+- **Lookalike package-source rejection** — trusted OpenClaw package sources now reject lookalike sibling paths (e.g., trusting `/artifactory/openclaw` no longer admits `/artifactory/openclaw-malicious`)
+
+### Mitigation
+
+```bash
+# Upgrade to v2026.6.8 or later (ideally the latest stable)
+npm update -g openclaw
+openclaw --version
+```
+
+---
+
 ## Gateway Authentication Bypass (JFrog)
 
 JFrog published "[Giving OpenClaw the Keys to Your Kingdom](https://jfrog.com/blog/giving-openclaw-the-keys-to-your-kingdom-read-this-first/)" revealing that **93.4% of publicly reachable OpenClaw instances had critical authentication bypass vulnerabilities**.
@@ -374,6 +406,9 @@ When these findings were reported to creator Peter Steinberger, his response was
 | Feb 9 | Follow-up: 135,000+ exposed instances, 42,665 on Shodan | **Critical** |
 | Feb 9 | JFrog: 93.4% of exposed instances have auth bypass | **Critical** |
 | Feb 9 | Gartner: "Unacceptable cybersecurity risk" published | — |
+| Jun 12 | v2026.6.6 released — security-hardening release (140+ PRs) | Mitigation |
+| Jun 30 | ~45 security advisories batch-published (MCP loopback, plugin-install bypasses, model-override auth) — most patched in v2026.6.6–v2026.6.8 | **High** |
+| Jun 30 | v2026.6.11 released — DOMPurify fix (GHSA-cmwh-pvxp-8882), lookalike package-source rejection | Mitigation |
 
 ---
 

docs/security/overview.md

@@ -119,7 +119,7 @@ All of this abuses intended capabilities — no software vulnerability required.
 
 For those who want to go deeper than the 5 quick steps:
 
-- [ ] **Update to the latest stable release** (currently v2026.6.1) — patches critical RCE and many subsequent security improvements
+- [ ] **Update to the latest stable release** (currently v2026.6.11) — patches critical RCE, the [June 30, 2026 advisory batch](/security/known-vulnerabilities#june-2026-advisory-batch-june-30-2026) (fixed in v2026.6.6–v2026.6.8), and many subsequent security improvements
 - [ ] **Bind gateway to localhost** — never expose port 18789
 - [ ] **Enable authentication** — token or password mode
 - [ ] **Set `trustedProxies`** — if behind any reverse proxy