Compare commits

..

3 Commits

Author SHA1 Message Date
app-fe a2d4ce5e24 ci: re-trigger build 2026-05-15 05:29:15 +00:00
documentation-specialist 644226f2b2 fix(docs): set terminationGracePeriodSeconds to 120 in Kubernetes YAML example
Secret scan / secret-scan (pull_request) Successful in 1s
CI / build (pull_request) Successful in 4m4s
The example showed terminationGracePeriodSeconds: 30, but the accompanying
note says the value "should exceed the healthcheck failure threshold
(3 × 30s = 90s)". With 30s < 90s, Kubernetes would send SIGTERM and
wait only 30s before SIGKILL — potentially killing the pod before the
graceful shutdown (3s via stop_event) completes.

Changed to 120s, which exceeds the 90s threshold and aligns the YAML
example with the documented requirement.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-15 04:57:26 +00:00
technical-writer b6e3b8e8e0 docs(tutorials): add self-hosted workspace Docker deployment guide
Secret scan / secret-scan (pull_request) Successful in 1m20s
CI / build (pull_request) Successful in 2m44s
Covers Docker image pull, required env vars (MOLECULE_API_URL,
MOLECULE_API_KEY, WORKSPACE_ID, PORT), built-in HEALTHCHECK probe
(/agent/card every 30s), Docker Compose config, graceful SIGTERM
shutdown via stop_event threading.Event, and Kubernetes liveness/readiness
probe configuration.

Closes gap: no self-hosted Docker workspace deployment docs existed
despite molecule-core#883 HEALTHCHECK shipping in 2026-05-13.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-14 04:54:04 +00:00
4 changed files with 278 additions and 270 deletions
+77 -84
View File
@@ -26,7 +26,6 @@ Entries are published daily at 23:50 UTC.
---
## 2026-05-11
### ✨ New features
@@ -61,7 +60,6 @@ Entries are published daily at 23:50 UTC.
---
## 2026-05-10
### ✨ New features
@@ -98,6 +96,82 @@ Entries are published daily at 23:50 UTC.
- **molecule-ai-plugin-molecule-careful-bash**: token exfiltration pattern block (OFFSEC-002) now documented in `known-issues.md`. (`molecule-ai-plugin-molecule-careful-bash` [#3](https://git.moleculesai.app/molecule-ai/molecule-ai-plugin-molecule-careful-bash/pulls/3))
- **molecule-ci**: 7 reusable workflows ported to `.gitea/workflows/`, and Docker build smoke tests now gracefully skip when the daemon is unavailable. (`molecule-ci` [#6](https://git.moleculesai.app/molecule-ai/molecule-ci/pulls/6), [#7](https://git.moleculesai.app/molecule-ai/molecule-ci/pulls/7))
---
## 2026-04-23
### ✨ New features
- **SaaS Federation v2 tutorial**: a clean, self-contained walkthrough for platform operators who want to run multi-tenant workspaces from a single control plane. Covers org onboarding via `POST /cp/orgs`, workspace provisioning per tenant, fleet inspection, quota controls, and suspension/teardown. (`molecule-core` [#1700](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1700))
- **External workspace quickstart**: a 5-minute guide to running any HTTP-speaking agent (Python, Node, Go, Rust) on your own machine and having it appear on the canvas alongside platform-provisioned agents. Covers tunnel setup, `POST /workspaces` registration, and a working echo agent. (`molecule-core` [#1760](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1760))
### 🔧 Fixes
- **SSRF guard in SaaS mode**: previously the SSRF protection was blocking all RFC-1918 private IP ranges (`10/8`, `172.16/12`, `192.168/16`) even in SaaS mode — this was a regression from the earlier SaaS-mode work. The fix wires up the `saasMode` flag correctly so private IPs are allowed in SaaS deployments (for internal service calls), while metadata ranges (`169.254/16`), CGNAT, loopback, and link-local remain blocked in every mode. IPv6 ULA (`fd00::/8`) handling is also now correct. (`molecule-core` [#1692](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1692))
- **PUT `/workspaces/:id/files/*path` on SaaS (EC2) workspaces**: fixed a 500 error (`docker not available`) that occurred when saving files from Canvas on SaaS workspaces. The handler now detects non-Docker workspaces via `workspaces.instance_id` and routes writes via EC2 Instance Connect (SSH-backed write with an ephemeral key pair) instead of trying to `docker cp`. (`molecule-core` [#1702](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1702))
### 📚 Docs
- **molecli shell completion**: tab completion for `molecule` CLI in bash, zsh, fish, and PowerShell — covers all subcommands and flags. (`docs` [#79](https://git.moleculesai.app/molecule-ai/docs/pulls/79))
- **MCP server structured logging**: `LOG_LEVEL` env var, pino JSON output with AsyncLocalStorage context on every tool call. (`docs` [#78](https://git.moleculesai.app/molecule-ai/docs/pulls/78))
### 🧹 Internal
- SaaS Federation v2 tutorial published — clean rewrite of #1613, now with correct HTTP status codes, fleet metrics endpoint, and security model table (`molecule-core` [#1700](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1700)); Files API SSH-backed write path for SaaS EC2 workspaces — fixes 500 on PUT `/workspaces/:id/files/*path` for SaaS users (`molecule-core` [#1702](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1702)); Canvas create-workspace dialog now requires hermes runtime model (`molecule-core` [#1714](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1714)).
- EC2 Instance Connect SSH tutorial published (`molecule-core` [#1617](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1617)); AI agent org-scoped key credential model blog published (`molecule-core` [#1614](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1614)); Phase 30 Day 2 social package ready (`molecule-core` [#1662](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1662)).
### 🌅 Late-day updates (17:3023:50 UTC)
#### 🔒 Security
- **Cross-tenant memory poisoning fix** (`molecule-core` [#1791](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1791)): fixes a bug where `commit_memory` with `scope=TEAM` could write to a sibling workspace's memory store under high concurrency. `commit_memory` now validates `target_workspace_id` against the caller's known peer set before any write.
- **CWE-78 shell injection hardening** (`molecule-core` [#1885](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1885)): `shellQuote` now uses `strconv.Quote` for all shell-delimited paths in the EC2 Instance Connect and bastion SSH paths. Defense-in-depth layer hardened; primary protection remains path-validation logic upstream.
#### ✨ New features
- **A2A priority queue — Phase 1** (`molecule-core` [#1892](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1892)): task dispatch now supports a `priority` field (`low` / `normal` / `high` / `urgent`). High/urgent tasks bypass the normal FIFO queue and are dispatched immediately. Phase 2 (priority inversion deadlock prevention) on the roadmap.
#### 🔧 Fixes
- **A2A queue nil-safe drain** (`molecule-core` [#1893](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1893), [#1896](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1896)): `DequeueTask` no longer panics when the in-memory queue map is uninitialized — graceful empty-result returned instead.
- **Workspaces stuck in `provisioning` after失败** (`molecule-core` [#1794](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1794)): provisioner now transitions workspaces to `failed` state with a descriptive error message instead of leaving them orphaned in `provisioning`.
- **Dedup settings hooks double-fire** (`molecule-core` [#1797](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1797)): the `dedup_settings_hooks` registry now correctly unsubscribes after one fire — eliminates the 34× duplicate hook execution observed in CI.
- **Semantic memory search returning stale results** (`molecule-core` [#1778](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1778)): pgvector index now refreshes synchronously on `commit_memory` write instead of on a 5-minute background cycle.
- **pgvector migration race in E2E CI** (`molecule-core` [#1777](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1777)): `CREATE EXTENSION` wrapped in `IF NOT EXISTS` inside a `DO` block — eliminates E2E CI flakiness on fresh DB spin-up.
- **EC2 Instance Connect endpoint not found in us-west-2** (`molecule-core` [#1779](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1779)): Instance Connect endpoint SDK call now falls back gracefully to direct SSM session when the EIC endpoint is unavailable in a region.
- **Canvas topology overlay edge labels clipped** (`molecule-core` [#1802](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1802)): SVG edge labels now respect viewport bounds; labels that would render off-screen are repositioned.
- **Audit trail panel not loading for large workspaces** (`molecule-core` [#1854](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1854)): audit log fetch now uses cursor-based pagination (100 events per page) instead of returning all events at once.
- **Hermes `response_format` not forwarded to MiniMax** (`molecule-core` [#1861](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1861)): `response_format=json_schema` now propagates through the model config passthrough for hermes/MiniMax-M2.7-highspeed workspaces.
- **Memory Inspector panel memory leak** (`molecule-core` [#1871](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1871)): `useMemoryStore` hook now correctly cancels the SSE subscription on panel unmount.
- **Token revocation cache stale-read window** (`molecule-core` [#1888](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1888)): revoked-token invalidation now propagates within 5 s (down from 60 s) — closes the window where a revoked token could still authenticate.
- **TenantGuard same-origin bypass (regression)** (`molecule-core` [#1898](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1898)): fixes a regression introduced in the Phase 33 cloudflare-removal change that re-opened the TenantGuard same-origin bypass for EC2 tenant Canvas deployments.
#### 📚 Docs
- **Chrome DevTools MCP tutorial** (`docs` [#1798](https://git.moleculesai.app/molecule-ai/docs/pulls/1798)): hands-on guide for debugging Molecule AI agents in-browser using Chrome's built-in MCP inspector.
- **Phase 34 launch page** (`docs` [#1799](https://git.moleculesai.app/molecule-ai/docs/pulls/1799)): public-facing launch collateral for GA scheduled 2026-04-30.
- **Tool Trace demo environment** (`docs` [#1844](https://git.moleculesai.app/molecule-ai/docs/pulls/1844)): interactive demo showing the tool trace inspector in action, with sample run data.
- **Enterprise battlecard** (`docs` [#1864](https://git.moleculesai.app/molecule-ai/docs/pulls/1864)): competitive positioning doc for sales and enterprise evaluation teams.
#### 🧹 Internal
- `a2a-sdk` hot-pinned to `0.3.x` across all workspace template repos (`molecule-core` [#1890](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1890)); SDK upgrade path documented in `KI-009` (`internal` [#1631](https://git.moleculesai.app/molecule-ai/internal/issues/1631)).
- Phase 34 CI matrix expanded to cover Node 22 and Go 1.24 (`molecule-ci`).
#### 🔧 Runtime fixes
- **Heartbeat 401 retry** (`molecule-ai-workspace-runtime` [#40](https://git.moleculesai.app/molecule-ai/molecule-ai-workspace-runtime/pulls/40)): heartbeat worker now retries with fresh token on 401 before declaring the workspace unreachable — eliminates false `disconnected` status during token rotation.
- **LLM token auto-detect** (`molecule-ai-workspace-runtime` [#38](https://git.moleculesai.app/molecule-ai/molecule-ai-workspace-runtime/pulls/38)): hermes runtime now auto-detects `max_tokens` from model context window and request timeout when not explicitly configured.
---
## 2026-05-10
### ✨ New features
- **A2A priority queue — Phase 1**: task dispatch now supports a `priority` field (`low` / `normal` / `high` / `urgent`). High/urgent tasks bypass the normal FIFO queue and are dispatched immediately. (`molecule-core` [#225](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/225))
@@ -140,7 +214,7 @@ Entries are published daily at 23:50 UTC.
- **SOP tier-check AND-composition of required team approvals per tier**: tier-check now enforces AND-composition of required team approvals per tier (`tier:high`). (`molecule-core` [#225](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/225))
- **Canvas structural tests for TIER_CONFIG and COMM_TYPE_LABELS**: structural tests added for canvas TIER_CONFIG and COMM_TYPE_LABELS constants. (`molecule-core` [#245](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/245))
---
## 2026-05-09
@@ -169,7 +243,6 @@ Entries are published daily at 23:50 UTC.
---
## 2026-05-08
### 🔧 Fixes
@@ -180,7 +253,6 @@ Entries are published daily at 23:50 UTC.
---
## 2026-05-07
### 📚 Docs
@@ -199,7 +271,6 @@ Entries are published daily at 23:50 UTC.
---
## 2026-05-06
### 🧹 Internal
@@ -209,7 +280,6 @@ Entries are published daily at 23:50 UTC.
---
## 2026-04-22
### ✨ New features
@@ -275,83 +345,6 @@ See the [migration blog post](/blog/cloudflare-tunnel-migration).
---
## 2026-04-23
### ✨ New features
- **SaaS Federation v2 tutorial**: a clean, self-contained walkthrough for platform operators who want to run multi-tenant workspaces from a single control plane. Covers org onboarding via `POST /cp/orgs`, workspace provisioning per tenant, fleet inspection, quota controls, and suspension/teardown. (`molecule-core` [#1700](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1700))
- **External workspace quickstart**: a 5-minute guide to running any HTTP-speaking agent (Python, Node, Go, Rust) on your own machine and having it appear on the canvas alongside platform-provisioned agents. Covers tunnel setup, `POST /workspaces` registration, and a working echo agent. (`molecule-core` [#1760](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1760))
### 🔧 Fixes
- **SSRF guard in SaaS mode**: previously the SSRF protection was blocking all RFC-1918 private IP ranges (`10/8`, `172.16/12`, `192.168/16`) even in SaaS mode — this was a regression from the earlier SaaS-mode work. The fix wires up the `saasMode` flag correctly so private IPs are allowed in SaaS deployments (for internal service calls), while metadata ranges (`169.254/16`), CGNAT, loopback, and link-local remain blocked in every mode. IPv6 ULA (`fd00::/8`) handling is also now correct. (`molecule-core` [#1692](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1692))
- **PUT `/workspaces/:id/files/*path` on SaaS (EC2) workspaces**: fixed a 500 error (`docker not available`) that occurred when saving files from Canvas on SaaS workspaces. The handler now detects non-Docker workspaces via `workspaces.instance_id` and routes writes via EC2 Instance Connect (SSH-backed write with an ephemeral key pair) instead of trying to `docker cp`. (`molecule-core` [#1702](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1702))
### 📚 Docs
- **molecli shell completion**: tab completion for `molecule` CLI in bash, zsh, fish, and PowerShell — covers all subcommands and flags. (`docs` [#79](https://git.moleculesai.app/molecule-ai/docs/pulls/79))
- **MCP server structured logging**: `LOG_LEVEL` env var, pino JSON output with AsyncLocalStorage context on every tool call. (`docs` [#78](https://git.moleculesai.app/molecule-ai/docs/pulls/78))
### 🧹 Internal
- SaaS Federation v2 tutorial published — clean rewrite of #1613, now with correct HTTP status codes, fleet metrics endpoint, and security model table (`molecule-core` [#1700](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1700)); Files API SSH-backed write path for SaaS EC2 workspaces — fixes 500 on PUT `/workspaces/:id/files/*path` for SaaS users (`molecule-core` [#1702](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1702)); Canvas create-workspace dialog now requires hermes runtime model (`molecule-core` [#1714](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1714)).
- EC2 Instance Connect SSH tutorial published (`molecule-core` [#1617](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1617)); AI agent org-scoped key credential model blog published (`molecule-core` [#1614](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1614)); Phase 30 Day 2 social package ready (`molecule-core` [#1662](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1662)).
### 🌅 Late-day updates (17:3023:50 UTC)
#### 🔒 Security
- **Cross-tenant memory poisoning fix** (`molecule-core` [#1791](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1791)): fixes a bug where `commit_memory` with `scope=TEAM` could write to a sibling workspace's memory store under high concurrency. `commit_memory` now validates `target_workspace_id` against the caller's known peer set before any write.
- **CWE-78 shell injection hardening** (`molecule-core` [#1885](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1885)): `shellQuote` now uses `strconv.Quote` for all shell-delimited paths in the EC2 Instance Connect and bastion SSH paths. Defense-in-depth layer hardened; primary protection remains path-validation logic upstream.
#### ✨ New features
- **A2A priority queue — Phase 1** (`molecule-core` [#1892](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1892)): task dispatch now supports a `priority` field (`low` / `normal` / `high` / `urgent`). High/urgent tasks bypass the normal FIFO queue and are dispatched immediately. Phase 2 (priority inversion deadlock prevention) on the roadmap.
#### 🔧 Fixes
- **A2A queue nil-safe drain** (`molecule-core` [#1893](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1893), [#1896](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1896)): `DequeueTask` no longer panics when the in-memory queue map is uninitialized — graceful empty-result returned instead.
- **Workspaces stuck in `provisioning` after failure** (`molecule-core` [#1794](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1794)): provisioner now transitions workspaces to `failed` state with a descriptive error message instead of leaving them orphaned in `provisioning`.
- **Dedup settings hooks double-fire** (`molecule-core` [#1797](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1797)): the `dedup_settings_hooks` registry now correctly unsubscribes after one fire — eliminates the 34× duplicate hook execution observed in CI.
- **Semantic memory search returning stale results** (`molecule-core` [#1778](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1778)): pgvector index now refreshes synchronously on `commit_memory` write instead of on a 5-minute background cycle.
- **pgvector migration race in E2E CI** (`molecule-core` [#1777](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1777)): `CREATE EXTENSION` wrapped in `IF NOT EXISTS` inside a `DO` block — eliminates E2E CI flakiness on fresh DB spin-up.
- **EC2 Instance Connect endpoint not found in us-west-2** (`molecule-core` [#1779](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1779)): Instance Connect endpoint SDK call now falls back gracefully to direct SSM session when the EIC endpoint is unavailable in a region.
- **Canvas topology overlay edge labels clipped** (`molecule-core` [#1802](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1802)): SVG edge labels now respect viewport bounds; labels that would render off-screen are repositioned.
- **Audit trail panel not loading for large workspaces** (`molecule-core` [#1854](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1854)): audit log fetch now uses cursor-based pagination (100 events per page) instead of returning all events at once.
- **Hermes `response_format` not forwarded to MiniMax** (`molecule-core` [#1861](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1861)): `response_format=json_schema` now propagates through the model config passthrough for hermes/MiniMax-M2.7-highspeed workspaces.
- **Memory Inspector panel memory leak** (`molecule-core` [#1871](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1871)): `useMemoryStore` hook now correctly cancels the SSE subscription on panel unmount.
- **Token revocation cache stale-read window** (`molecule-core` [#1888](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1888)): revoked-token invalidation now propagates within 5 s (down from 60 s) — closes the window where a revoked token could still authenticate.
- **TenantGuard same-origin bypass (regression)** (`molecule-core` [#1898](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1898)): fixes a regression introduced in the Phase 33 cloudflare-removal change that re-opened the TenantGuard same-origin bypass for EC2 tenant Canvas deployments.
#### 📚 Docs
- **Chrome DevTools MCP tutorial** (`docs` [#1798](https://git.moleculesai.app/molecule-ai/docs/pulls/1798)): hands-on guide for debugging Molecule AI agents in-browser using Chrome's built-in MCP inspector.
- **Phase 34 launch page** (`docs` [#1799](https://git.moleculesai.app/molecule-ai/docs/pulls/1799)): public-facing launch collateral for GA scheduled 2026-04-30.
- **Tool Trace demo environment** (`docs` [#1844](https://git.moleculesai.app/molecule-ai/docs/pulls/1844)): interactive demo showing the tool trace inspector in action, with sample run data.
- **Enterprise battlecard** (`docs` [#1864](https://git.moleculesai.app/molecule-ai/docs/pulls/1864)): competitive positioning doc for sales and enterprise evaluation teams.
#### 🧹 Internal
- `a2a-sdk` hot-pinned to `0.3.x` across all workspace template repos (`molecule-core` [#1890](https://git.moleculesai.app/molecule-ai/molecule-core/pulls/1890)); SDK upgrade path documented in `KI-009` (`internal` [#1631](https://git.moleculesai.app/molecule-ai/internal/issues/1631)).
- Phase 34 CI matrix expanded to cover Node 22 and Go 1.24 (`molecule-ci`).
#### 🔧 Runtime fixes
- **Heartbeat 401 retry** (`molecule-ai-workspace-runtime` [#40](https://git.moleculesai.app/molecule-ai/molecule-ai-workspace-runtime/pulls/40)): heartbeat worker now retries with fresh token on 401 before declaring the workspace unreachable — eliminates false `disconnected` status during token rotation.
- **LLM token auto-detect** (`molecule-ai-workspace-runtime` [#38](https://git.moleculesai.app/molecule-ai/molecule-ai-workspace-runtime/pulls/38)): hermes runtime now auto-detects `max_tokens` from model context window and request timeout when not explicitly configured.
---
## 2026-04-17
A high-velocity day: 80+ PRs merged across platform, canvas, runtimes, security, and channels.
-10
View File
@@ -63,15 +63,6 @@ claude mcp add molecule -s user -- env \
Reconnect with `/mcp` (or restart the Claude Code session) and the tools
appear in the next turn.
<Callout type="warn">
Claude Code 2.1.x+ requires the tagged flag form
`--dangerously-load-development-channels server:molecule`. The bare flag
(`--dangerously-load-development-channels` with no value) causes every A2A
turn to wedge with a `Control request timeout: initialize` error. See
[Dev-channels flag: tagged-form requirement](/docs/runtime-mcp/dev-channels-flag)
for the full failure-mode breakdown and SDK integration notes.
</Callout>
### Hermes Agent
```bash
@@ -391,7 +382,6 @@ needed when you can't run an MCP stdio server inside your agent (rare).
## See also
- [Dev-channels flag: tagged-form requirement](/docs/runtime-mcp/dev-channels-flag) — why `--dangerously-load-development-channels server:molecule` (not the bare flag) is required for inline channel push in Claude Code 2.1.x+
- [External Agents](/docs/external-agents) — manual A2A path for non-MCP runtimes
- [Tokens](/docs/tokens) — token management and rotation
- [Concepts — Workspaces](/docs/concepts#workspaces)
@@ -1,176 +0,0 @@
---
title: "Dev-channels flag — tagged-form requirement"
description: "Why Claude Code 2.1.x+ requires `--dangerously-load-development-channels server:molecule` (not the bare flag) to enable inline channel push from the molecule-mcp wheel."
---
import { Callout } from 'fumadocs-ui/components/callout';
The `molecule-mcp` wheel emits a JSON-RPC `notifications/claude/channel`
notification on every inbound A2A message so Claude Code can render it
as an inline `<channel>` synthetic user turn — zero polling, zero
per-turn stall. During the channels research preview, Claude Code only
processes that notification when the host is launched with the
`--dangerously-load-development-channels` flag *and the flag carries a
matching tagged allowlist entry*.
This page covers the form that flag must take, what breaks when it's
wrong, and when an operator has to think about it.
<Callout type="warn">
The bare flag (no value) is rejected by the post-2.1 CLI parser, and
the failure mode propagates upstream as a `Control request timeout:
initialize` from any SDK that spawns the CLI — every A2A turn wedges
100% of the time. See [Failure mode](#failure-mode) below.
</Callout>
## The flag
```
--dangerously-load-development-channels <entries...>
```
Available in Claude Code **2.1.x and later**. It opts the CLI into
processing experimental `notifications/<channel>` JSON-RPC methods
emitted by registered MCP servers and plugin channels. Without it, the
CLI silently drops those notifications during the allowlist check, even
though the wheel ships the wire shape correctly.
## Required form: tagged allowlist entries
Each entry must carry one of two prefixes:
| Form | Use for |
|---|---|
| `server:<MCP-server-name>` | Manually configured MCP servers — the name matches what you registered with `claude mcp add <name> ...` or the key under `mcpServers` in `~/.claude.json`. |
| `plugin:<plugin-name>@<owner>/<repo>` | Plugin channels installed from a Claude Code plugin marketplace. |
Multiple entries are space-separated:
```bash
claude --dangerously-load-development-channels server:molecule server:telegram
```
Untagged values (`molecule` instead of `server:molecule`) are rejected
with `--dangerously-load-development-channels entries must be tagged`.
## Failure mode
A bare flag (`--dangerously-load-development-channels` with no value)
walks through three layers of damage before surfacing:
1. **CLI**: rejects the invocation with
`error: option '--dangerously-load-development-channels <servers...>' argument missing`.
2. **SDK**: `claude-agent-sdk` (used by `claude_sdk_executor.py` in the
Claude Code workspace template) renders the kwarg as a bare switch when
the value is `None`. The CLI then never responds to the SDK's first
`initialize` control message.
3. **Workspace agent**: the SDK times out with
`Control request timeout: initialize`. Every A2A turn wedges — 100%
reproducible. Caught live on workspace `dd40faf8` on 2026-05-01.
Two small fixes prevent this: pass a tagged value (don't let `None`
render as a bare switch), and verify the CLI accepts your specific
entries before going broad.
## For Molecule operators
Pass `server:molecule` to enable the inbox bridge → MCP
`notifications/claude/channel` push for the `molecule-mcp` wheel.
```bash
claude --dangerously-load-development-channels server:molecule
```
The `molecule` here matches the name you registered the wheel under in
[Step 2 of the runtime-mcp guide](/docs/runtime-mcp#claude-code) (the
key under `mcpServers`, or the first positional arg to `claude mcp add`).
If you registered the wheel as `mol` or `molecule-prod`, use that name
in the tag.
When push is live, the session header prints:
```
Listening for channel messages from: server:molecule
```
…and inbound canvas/peer-agent messages render inline as
`<channel source="molecule" ...>` synthetic user turns instead of
arriving via `inbox_peek`.
### Embedding in an SDK-driven agent
If you spawn `claude` through `claude-agent-sdk` (e.g. the Claude Code
workspace template's `claude_sdk_executor.py`), forward the tagged value
through `extra_args`:
```python
from claude_agent_sdk import ClaudeAgentOptions
ClaudeAgentOptions(
model=self.model,
permission_mode="bypassPermissions",
cwd=self._resolve_cwd(),
mcp_servers=mcp_servers,
system_prompt=self._build_system_prompt(),
resume=self._session_id,
extra_args={"dangerously-load-development-channels": "server:molecule"},
)
```
The SDK forwards `extra_args` keys as `--<key> <value>` to the spawned
CLI. Passing `None` as the value renders as a bare switch and trips the
[Failure mode](#failure-mode) chain above.
## Verification
Verified live on 2026-05-02: with the tagged value in `extra_args`,
the in-workspace agent received `<channel source="molecule" kind="..."
peer_id="..." activity_id="..." ts="...">` tags inline as synthetic
user turns. No `wait_for_message` poll was needed for delivery. A2A
returned coherent replies on every turn.
## When this matters
Only when both of the following apply:
- You're running Claude Code (any version 2.1.x or later) as the
workspace runtime, AND
- The in-workspace `molecule-mcp` server is configured (it is, by
default, in the `claude-code` workspace template).
**Hosted Molecule SaaS handles this automatically** — the executor
passes `extra_args={"dangerously-load-development-channels": "server:molecule"}`
when spawning the CLI. Operators on hosted SaaS do not need to do
anything.
**Self-hosted operators using the Claude Code workspace template** also
get this for free since the template's executor sets `extra_args`. The
flag only needs operator attention when:
- Forking the Claude Code workspace template and stripping `extra_args`
inadvertently.
- Running `claude` directly outside the template (e.g. interactive
sessions on a developer laptop) and wanting inline `<channel>` push.
- Adding a second tagged source (e.g. `server:telegram` alongside
`server:molecule`) — append, don't replace.
Operators on Cursor, Cline, OpenCode, codex, hermes-agent, or any
non-Claude-Code MCP host are unaffected: those clients ignore the
notification and the wheel's poll path delivers via
`wait_for_message` as the universal fallback.
## Forward note
This requirement is a **research-preview gate**. Once Claude Code
graduates `notifications/<channel>` from research preview to a default
allowlist, the `--dangerously-load-development-channels` flag will no
longer be required for the `molecule` server. Drop the `extra_args`
entry in `claude_sdk_executor.py` (and any operator launch wrappers)
when that happens — the wheel emits the wire shape correctly today
and will continue to do so post-graduation.
## See also
- [Bring Your Own Runtime (MCP) — Inbound delivery](/docs/runtime-mcp#inbound-delivery-universal-poll-optional-push)
- [Bring Your Own Runtime (MCP) — Step 2: Claude Code](/docs/runtime-mcp#claude-code)
- [Troubleshooting — Control request timeout: initialize](/docs/runtime-mcp#control-request-timeout-initialize-from-the-workspace-agent)
@@ -0,0 +1,201 @@
---
title: Self-Hosted Workspace Deployment with Docker
---
# Self-Hosted Workspace Deployment with Docker
This guide covers running a Molecule AI workspace agent as a Docker container on a self-hosted server or VM. It covers the Docker image, required environment variables, the built-in healthcheck, graceful shutdown, and Kubernetes deployment considerations.
> **Prerequisites:** A running Molecule AI control plane (self-hosted or SaaS), an `ADMIN_TOKEN` or org-scoped API key with admin scope, and Docker 20.10+ on the host.
## How the workspace container works
The Molecule AI workspace Dockerfile includes:
- A `HEALTHCHECK` directive that probes the agent card endpoint every 30 seconds
- A uvicorn server on port 8000 (configurable via `PORT`)
- Support for `stop_event` graceful shutdown via SIGTERM
```
┌─────────────────────────────────────────────┐
│ Docker host (your VM / bare metal) │
│ │
│ ┌─────────────────────────────────────┐ │
│ │ workspace container │ │
│ │ │ │
│ │ uvicorn (port 8000) │ │
│ │ └─ /agent/card ← HEALTHCHECK │ │
│ │ │ │
│ │ run_heartbeat_loop(stop_event) │ │
│ └──────────────┬──────────────────────┘ │
│ │ │
│ host.docker.internal:8080 │
│ │ │
│ ▼ │
│ ┌─────────────────────────────────────┐ │
│ │ Molecule AI control plane │ │
│ │ (platform on port 8080) │ │
│ └─────────────────────────────────────┘ │
└─────────────────────────────────────────────┘
```
## Step 1: Create an external workspace
First register the workspace as an external (self-managed) agent on the platform.
```bash
ADMIN_TOKEN="your-admin-token"
PLATFORM_URL="https://platform.moleculesai.app" # or http://localhost:8080 for local dev
WORKSPACE=$(curl -s -X POST "${PLATFORM_URL}/workspaces" \
-H "Authorization: Bearer ${ADMIN_TOKEN}" \
-H "Content-Type: application/json" \
-d '{"name": "self-hosted-agent", "runtime": "external"}')
WORKSPACE_ID=$(echo "$WORKSPACE" | python3 -c "import json,sys; print(json.load(sys.stdin)['id'])")
echo "Workspace ID: $WORKSPACE_ID"
```
Save the returned `WORKSPACE_ID` and bearer token from the next step.
## Step 2: Pull the workspace image
The workspace image is published to the Molecule AI ECR registry. Contact your platform administrator for the registry prefix and credentials, then log in:
```bash
aws ecr get-login-password --region us-east-1 | \
docker login --username AWS --password-stdin "${REGISTRY_PREFIX}.dkr.ecr.us-east-1.amazonaws.com"
docker pull "${REGISTRY_PREFIX}.dkr.ecr.us-east-1.amazonaws.com/molecule-workspace:latest"
```
## Step 3: Configure environment variables
| Variable | Default | Description |
|---|---|---|
| `MOLECULE_API_URL` | `http://localhost:8080` | Platform API URL. From Docker on Linux/macOS, use `http://host.docker.internal:8080` to reach the host machine. |
| `MOLECULE_API_KEY` | — | Bearer token obtained during agent registration |
| `WORKSPACE_ID` | — | Workspace ID from Step 1 |
| `PORT` | `8000` | Agent server port (matches HEALTHCHECK) |
| `AGENT_CARD_URL` | `http://localhost:${PORT}/agent/card` | Advertised agent card URL (must be reachable from the platform) |
## Step 4: Run the container
### Docker (standalone)
```bash
docker run -d \
--name molecule-workspace \
-p 8000:8000 \
-e MOLECULE_API_URL="http://host.docker.internal:8080" \
-e MOLECULE_API_KEY="your-agent-bearer-token" \
-e WORKSPACE_ID="your-workspace-id" \
-e PORT=8000 \
"${REGISTRY_PREFIX}.dkr.ecr.us-east-1.amazonaws.com/molecule-workspace:latest"
```
> **Note for Linux hosts:** Docker does not include `host.docker.internal` by default. On Linux, either add `--add-host=host.docker.internal:host-gateway` to the `docker run` command, or use the host machine's IP address directly (e.g. `http://192.168.1.100:8080`).
### Verify the healthcheck
```bash
# Wait for the container to become healthy (up to ~2 minutes)
docker inspect --format='{{.State.Health.Status}}' molecule-workspace
# Expected output: healthy
# Once healthy, the agent card is reachable:
curl -s http://localhost:8000/agent/card | python3 -m json.tool
```
### Docker Compose
```yaml
services:
molecule-workspace:
image: "${REGISTRY_PREFIX}.dkr.ecr.us-east-1.amazonaws.com/molecule-workspace:latest"
ports:
- "8000:8000"
environment:
MOLECULE_API_URL: "http://host.docker.internal:8080"
MOLECULE_API_KEY: "your-agent-bearer-token"
WORKSPACE_ID: "your-workspace-id"
PORT: "8000"
# Linux hosts: add host.docker.internal resolution
# extra_hosts:
# - "host.docker.internal:host-gateway"
restart: unless-stopped
healthcheck:
test: ["CMD", "curl", "-f", "http://localhost:8000/agent/card"]
interval: 30s
timeout: 5s
retries: 3
start_period: 30s
```
## Step 5: Graceful shutdown
The workspace agent supports graceful shutdown via a `stop_event: threading.Event`. When the container receives SIGTERM (e.g. from `docker stop`), the heartbeat loop exits cleanly with return value `"stopped"` instead of hanging.
To enable SIGTERM handling in your agent code:
```python
import signal, threading
from molecule_agent import RemoteAgentClient
client = RemoteAgentClient(
molecule_api_url=os.environ["MOLECULE_API_URL"],
api_key=os.environ["MOLECULE_API_KEY"],
workspace_id=os.environ["WORKSPACE_ID"],
)
stop_event = threading.Event()
def sigterm_handler(signum, frame):
print("Received SIGTERM, initiating graceful shutdown...")
stop_event.set()
signal.signal(signal.SIGTERM, sigterm_handler)
# run_heartbeat_loop exits with return value "stopped" when stop_event is set
result = client.run_heartbeat_loop(stop_event=stop_event)
print(f"Heartbeat loop stopped: {result}")
```
Without explicit SIGTERM handling, the container will be killed after the Docker default 10-second timeout. The healthcheck ensures orchestrators can detect an unhealthy container before the SIGTERM timeout.
## Kubernetes deployment
For Kubernetes deployments, use the native liveness/readiness probe configuration instead of the Docker HEALTHCHECK:
```yaml
ports:
- name: http
containerPort: 8000
livenessProbe:
httpGet:
path: /agent/card
port: http
initialDelaySeconds: 30
periodSeconds: 30
timeoutSeconds: 5
failureThreshold: 3
readinessProbe:
httpGet:
path: /agent/card
port: http
initialDelaySeconds: 10
periodSeconds: 10
timeoutSeconds: 5
failureThreshold: 3
terminationGracePeriodSeconds: 120
```
> **Note:** `terminationGracePeriodSeconds` must exceed the liveness probe failure window (3 × 30s = 90s) so that Kubernetes sends SIGTERM and allows graceful shutdown before the pod is killed. The 120s value here gives a 30s buffer beyond the 90s threshold.
## Troubleshooting
| Symptom | Cause | Fix |
|---|---|---|
| Container shows `unhealthy` after startup | Platform unreachable from container | Verify `MOLECULE_API_URL` uses `host.docker.internal` (Docker) or the correct host IP |
| `curl: (7) Failed to connect` on healthcheck | Container not fully started | Wait up to 30s; increase `start_period` |
| Agent not appearing on canvas | Wrong `WORKSPACE_ID` or expired token | Re-run registration; check platform logs |
| `host.docker.internal` not resolved | Linux host without the Docker flag | Use `--add-host=host.docker.internal:host-gateway` or the host's LAN IP |