Changelog

Versioning note

This log runs reverse-chronologically. The version scheme was reshuffled once:

From 2026-05 onward: gradual v0.6 → v0.7 → v0.8 → v0.9 → v0.10 releases; the v0.X.Y format mirrors commhub-server's 0.X.Y semver style.
Before 2026-04: used v1.0.0-preview.N / v2.1 style version numbers that overpromised. Deprecated.
Current stable: v0.10.13 (2026-06-08, shipped via npm latest tag; v0.8.1 was the first Apache 2.0 OSS release).
Older entries kept for git-blame continuity — see v1.0.0-preview / v2.1 / v0.x sections below.

v0.10.13 — `grok-build-acp` `session/prompt` 300s timeout hang fix (P0 hotfix) (2026-06-08) ✅ stable

Version alignment (npm latest tag):

@sleep2agi/agent-node@2.4.9 ← bumped (#210 / #204 runtime — ACP handleServerRequest non-integer error-code coerce fix)
@sleep2agi/agent-network@2.2.10 ← unchanged
@sleep2agi/commhub-server@0.8.4 ← unchanged (PINNED)

🌟 Highlights

Root cause + fix for `grok-build-acp` nodes hanging at `session/prompt timed out after 300000ms`

Symptom: A grok-build-acp runtime node accepts a second task, hangs for ≈5 min, then agent-node reports grok ACP request 'session/prompt' timed out after 300000ms. ai-insight's A站Grok node (grok 0.2.29 alpha) captured the exact 2026-06-07 19:53:09 log:

text

ERROR failed to parse incoming message: invalid type: string 'ENOENT',
expected i32 at line 1 column 48.
Raw: {'jsonrpc':'2.0','id':5,
      'error':{'code':'ENOENT',
               'message':'ENOENT: no such file or directory, open ...'}}

Root cause: ACP server-request responses (e.g. a failed read_file) carry a JS-native string error code like 'ENOENT', but the Grok agent's protocol requires code to be an i32 integer. The old agent-node passed the string straight through → Grok agent parse failure → undefined state → hang until the client-side 300 s timeout.

Fix (commit 4818776): client.ts:handleServerRequest adds a Number.isInteger(rawCode) guard. Non-integer codes are coerced to -32000 (the JSON-RPC standard reserved range); the original string is preserved on data.originalCode so no information is lost.

New regression tests: +2 cases, bun test 89/89 pass.

Live verification (in-house, 2026-06-07 19:50–19:55):

Same-shape read_file failure retried: returns structured code: -32000 + data.originalCode: "ENOENT" immediately
The grok turn continues without hanging; the task completes normally (47 s, well under the 300 s timeout)
ai-insight's A站Grok node passed UAT after installing 2.4.9-preview.0 globally

🐛 Bugs Fixed

#210 / #204 runtime — grok-build-acp ACP server-request responses carrying non-integer error codes (e.g. ENOENT) caused the Grok agent to fail to parse and hang until the 300 s timeout

📦 Install (fresh install)

bash

npm i -g @sleep2agi/agent-network@latest @sleep2agi/agent-node@latest
# Verify versions
anet --version          # agent-network v2.2.10 (unchanged)
agent-node --version    # agent-node v2.4.9 ⬆

anet hub start auto-fetches commhub-server@0.8.4 (PINNED, unchanged).

🔄 Upgrade (existing users)

Narrow path (recommended — only this package needs the hotfix):

bash

npm i -g @sleep2agi/agent-node@2.4.9
# Restart every grok-build-acp node
cd <your-anet-workdir>
anet node stop <grok-node-alias> && anet node start <grok-node-alias>

Full sweep (also refreshes READMEs / metadata):

bash

anet upgrade

⚠️ Node version note: agent-node supports Node ≥ 18; this hotfix's Docker smoke ran clean on both Node 20.20 and 24.16.

For the full troubleshooting entry see troubleshooting → grok-build-acp node task hangs.

🙏 Credits

Bug repro + root cause + UAT: live 19:53:09 capture + Vincent's A站Grok 47 s pilot; fix implementation: commit 4818776 + 2 regression tests; release ops: Method B two-phase + dual Install/Upgrade release notes sections.

Full Changelog: https://github.com/sleep2agi/agent-network/compare/v0.10.12...v0.10.13

v0.10.12 — Grok-build runtime scenario enablement + 0.2.8 alpha regression verify (2026-05-30) ✅ stable

Version alignment (npm latest tag):

@sleep2agi/agent-node@2.4.8 ← bumped (scenario docs + 0.2.8 alpha regression baseline tag)
@sleep2agi/agent-network@2.2.10 ← unchanged
@sleep2agi/commhub-server@0.8.4 ← unchanged

Highlights

Video-generation scenario (0 LOC): Send a Grok node a task carrying an image URL; the backend auto-routes to the grok-imagine-video model and returns mp4. Anet itself ships zero code changes — verified via ffmpeg first-frame visual check. See research/grok-video-gen-capability-probe.md (with Erratum) + scenarios/video-gen-marketing.en.md.
Basic X search (out-of-the-box): Find X URL + title + excerpt by keyword / handle — the LLM does it via web_search + allowed_domains=["x.com"]. No pre-setup required.
Live X advanced search (workspace setup required): Live streaming + per-post faves / retweets / replies metadata + since: / min_faves: operators — requires the user to pre-stage a twitterapi.io API key + fetcher script that the LLM drives via run_terminal_command. See scenarios/x-search-informant.en.md.
0.2.8 alpha regression verify: 87 / 87 bun unit tests pass; the #201 delegation detection + #204 .mcp.json isolation fixes hold on Grok 0.2.8 alpha. Detail: tests/p-grok-028-regression-verify/report.md.
RFC-021 §12 / §13: ACP capability dossier updated with Path D (workspace setup + terminal bypass) + XSearch ACP exposure test (XSearch's backend tool is structurally not exposed via ACP on 0.1.219 → 0.2.12 alpha; long-term resolution sits in the upstream xAI PR #1302).

Install / Upgrade

Non-breaking; 0.2.8 alpha regression verified, safe to upgrade:

bash

anet upgrade

Or single-package: npm i -g @sleep2agi/agent-node@2.4.8

Full Changelog: https://github.com/sleep2agi/agent-network/compare/v0.10.11...v0.10.12

v0.10.11 — #204 grok-build-acp per-node identity isolation + #194 commhub broadcast attribution hotfix (2026-05-28) ✅ stable

Version alignment (npm latest tag):

@sleep2agi/agent-network@2.2.10 ← bumped (anet hub stop / anet hub status subcommands #200 + anet hub start stderr inherit #199 silent-hang fix + PINNED commhub-server 0.8.3 → 0.8.4)
@sleep2agi/agent-node@2.4.7 ← bumped (#204 grok-build-acp per-node .anet/nodes/<alias>/runtime-cwd/ isolation + #201 Grok delegate parser 3-layer broaden)
@sleep2agi/commhub-server@0.8.4 ← bumped (#194 broadcast channel_meta_json sender attribution hotfix — from_session injection no longer overrides the real LLM agent alias)
@sleep2agi/agent-network-dashboard@0.5.6 ← unchanged

🌟 Highlights

#204 — grok-build-acp per-node isolated cwd

Problem: grok-build-acp runtime nodes shared the .mcp.json discovery path, causing stale .mcp.json identity pollution — if a node's working directory carried an old .mcp.json, a newly started Grok node would be misidentified as that prior node.

Fix: Every node forks an isolated cwd (.anet/nodes/<alias>/runtime-cwd/), decoupled from the discovery path. This eliminates stale .mcp.json interference.

E2E verification: cross-node dispatch on a live Grok node — channel sender attribution correctly attributed to the sending node's alias, proving the LLM-layer attribution path is uncontaminated.

#194 — commhub broadcast sender attribution hotfix

Problem: On cross-node broadcast, the channel_meta_json sender field went through the from_session injection path, which overrode the real LLM agent name.

Fix: commhub-server 0.8.4 corrects the from-name injection logic to preserve the real sender alias.

🐛 Bugs Fixed

#199 — anet hub start silent-hang fix: spawn stdio changed from "pipe" to "inherit" — commhub-server bunx fetch failures are now immediately visible.
#200 — anet hub stop / anet hub status subcommands: users no longer need manual lsof + kill. Now anet hub stop [--port <p>] (SIGTERM → 3s grace → SIGKILL) + anet hub status (PID + port + /health version).
#201 — Grok runtime refusing to delegate: explicit delegation parser 3-layer wrapper broaden + prompt softening covers all cases.

📦 Install (fresh install)

bash

npm i -g @sleep2agi/agent-network@latest
# Verify version
anet -v  # Should show v2.2.10

anet hub start auto-fetches commhub-server@0.8.4 (PINNED) + the first node start auto-fetches agent-node@2.4.7.

🔄 Upgrade (existing users)

bash

anet upgrade
# Or manually
npm i -g @sleep2agi/agent-network@latest

anet upgrade syncs agent-network + agent-node + commhub-server to the latest latest versions.

🙏 Credits

Shipped by the Agent Network team — design + lead review / agent-node #204 fix / agent-network release ops + commhub-server promote / testing + docs delivered end-to-end. Individual contribution detail on the v0.10.11 GitHub release page. LLM E2E attribution was verified on a live node via cross-hub dispatch.

Full Changelog: https://github.com/sleep2agi/agent-network/compare/v0.10.10...v0.10.11

v0.10.10 — Xiaomi MiMo full 5-model support + envRef wizard-to-start auto-link (2026-05-27) ✅ stable

Version alignment (npm latest tag):

@sleep2agi/agent-network@2.2.9 ← bumped (envRef Option A auto-source + anet -v now shows full prerelease suffix + Grok delegation parser broadened)
@sleep2agi/agent-node@2.4.6 ← bumped (envRef Option A implementation + Grok runtime stabilization continued)
@sleep2agi/commhub-server@0.8.3 ← unchanged
@sleep2agi/agent-network-dashboard ← unchanged

P0 — Xiaomi MiMo Vendor Preset complete

Added mimo-v2.5-tts-voicedesign — the full official 5-model lineup is now in the picker
anet node create → pick claude-agent-sdk runtime → "小米 MiMo" vendor → choose from 5 models: mimo-v2.5-pro (default) / mimo-v2.5 / mimo-v2-pro / mimo-v2-omni / mimo-v2.5-tts-voicedesign
Endpoint https://token-plan-cn.xiaomimimo.com/anthropic speaks the Anthropic Messages protocol
📌 Note: voicedesign is a TTS speech-design model — Anthropic Messages text requests are likely vendor-unsupported. For text dialogue, use the first 4 models.

P0 — envRef wizard-to-start auto-link (#193)

Pain point: After anet node create, running anet node start in the same shell reported FATAL env var not set — you had to manually export ANTHROPIC_AUTH_TOKEN_N_<id>=... first.

New behavior:

During wizard create, the API key is written to .anet/nodes/<alias>/.env (mode 0600, auto-added to .anet/.gitignore)
anet node start sources that .env file at launch — no manual export needed
Cross-machine deployment still supported (the wizard still prints the export command once so it can be copied to another box)
Debug logs emit only loaded N key(s) from .anet/nodes/<alias>/.env — the key value is never echoed
Backward compatible: existing ANTHROPIC_AUTH_TOKEN_N_* shell exports keep working; legacy plain config.json mode also still works

Applies to every claude-agent-sdk node (MiMo / MiniMax / InternLM / GLM / any Anthropic-Messages-compatible vendor).

Bug fixes

#192 anet -v no longer truncates the prerelease suffix — it now shows the full version (e.g. v2.2.9, including any -preview.N suffix when applicable)
#189 Grok runtime grok-build-acp fully stabilized: the explicit-delegation parser was broadened to cover cases like "你和 X 沟通一下…" / "send_task X 一下…" (no-punctuation trailing body), so cross-node delegation routing is reliable.

Known Issues

The internal npx fallback inside agent-network still pins @sleep2agi/agent-node@preview — a future preview push could expose @latest users to an unstable build. There's no regression in this release (preview 2.4.6-preview.2 is content-equivalent to latest 2.4.6), but follow-up handling is queued under a future RFC (tracking issue to be opened; distinct from the same-numbered RFC-021 ACP capability profile expansion which targets X-search unlock).

v0.10.9 — Dashboard image sending + CommHub attachment metadata + codex-sdk image input (2026-05-25) ✅ stable

Version alignment (npm latest tag):

@sleep2agi/agent-network@2.2.7 ← bumped (pinned server moved to @sleep2agi/commhub-server@0.8.3)
@sleep2agi/agent-node@2.4.3 ← bumped (codex-sdk runtime consumes structured image attachments)
@sleep2agi/commhub-server@0.8.3 ← bumped (meta_json persistence + MCP/REST attachment metadata)
@sleep2agi/agent-network-dashboard@0.5.4 ← bumped (TaskChatPanel image upload/paste send path)

P0 — Dashboard command flow can send images

Vincent catch: Dashboard could upload and preview images, but tasks sent to agents only carried text paths. The codex-sdk runtime never received actual image input, so screenshots, design drafts, and error images could not be dispatched from the web/mobile command surface.

Implementation:

Dashboard sends structured attachments for uploaded/pasted images while keeping text fallback paths and preview URLs.
Hub send_task and REST /api/task accept meta.attachments, persist it to inbox.meta_json and tasks.meta_json, and return parsed meta from get_inbox.
agent-node extracts local image paths from meta.attachments and passes them into codex-sdk/studio image input.
Telegram image dispatch now passes image arguments in the correct position.

Rollout + Smoke

Local CommHub and Dashboard were upgraded and restarted.
27 host codex-sdk agent-node processes were rolled to the new code.
P0 smoke sent /tmp/anet-image-smoke.png to 通信测试牛; node logs showed +1 image(s) / → processing [codex] +1 image(s), with reply 图片通道OK.

Known Limits

Dashboard uploads currently hand off local file paths, best suited for same-host hub/agent deployments. Cross-host object-storage delivery remains future work.
Existing old containers need reinstall/restart to pick up this release.
This release fixes delivery into the runtime; rich media history and mobile IM polish remain dashboard follow-up work.

See the v0.10.9 tag.

v0.10.8 — Dashboard Servers panel UI copy fix + TopoGraph density-tier polish (2026-05-17) ✅ stable

Version alignment (npm latest tag):

@sleep2agi/agent-network@2.2.6 (unchanged, v0.10.7)
@sleep2agi/agent-node@2.4.2 (unchanged)
@sleep2agi/commhub-server@0.8.2 (unchanged)
@sleep2agi/agent-network-dashboard@0.5.3 ← bumped (UI copy + Playwright attrs + polish fold-in)

Fix — #157 Dashboard Servers panel UI copy fix (Root cause #1)

Vincent caught this in real testing (with dashboard screenshot): the Servers panel was rendering agent rollup pending hub ≥ 0.8.2-preview / disk metric pending hub ≥ 0.8.2-preview for every hub, even though every production hub is already at ≥ 0.8.2. The copy was outdated and misleading — for a moment Vincent thought the version dashboard data was completely broken.

Root cause #1 (this patch): ServersDrawer UI carried placeholder text introduced during the 0.8.2 upgrade window. 0.8.2 has long since shipped, but the placeholder text was never removed, so it kept rendering "pending" for ≥ 0.8.2 hubs and led users to believe the hub data was missing.

Implementation (app/components/ServersDrawer.tsx):

diff

- <div ...>agent rollup pending hub ≥ 0.8.2-preview</div>
- <div ...>disk metric pending hub ≥ 0.8.2-preview</div>
+ <div ... data-server-agents-missing="true">agent rollup not reported by hub</div>
+ <div ... data-server-disk-missing="true">disk metric not reported by hub</div>

The new copy accurately says "this hub didn't report it right now" instead of implying the hub version is too old. The new data-server-agents-missing / data-server-disk-missing Playwright hooks enable the next e2e round to validate hub-side telemetry coverage.

Root causes #2 + #3 located, deferred

#2 (v0.10.9 candidate): missing dedupe when one hostname appears multiple times can double-count servers — the dashboard team's fix is queued for v0.10.9 ship.
#3 (v0.11.0 candidate): status=offline vs telemetry mismatch (telemetry still reports but SSE last_seen has timed out) — needs system-level status reconciliation.

Polish fold-in — TopoGraph density-tier (purely additive)

Dashboard team commit 3f73810 (0.5.3-preview.16) — the canvas state attribute data-topo-fleet-density-tier ∈ {empty, sparse, normal, dense, very-dense} exposes a 12th observable testing surface. Tier boundaries (sparse 1-3 / normal 4-15 / dense 16-30 / very-dense 31+) line up with the dense-layout collapse gate. Purely additive, no UX change. Paired with the numeric counts, e2e selectors now have a complete canvas-state snapshot.

Quality gates + lessons

Source-grep verify: the lead's grep -rh "not reported by hub" hits + the legacy copy only survives in JSDoc comments ✅
Docker preview smoke: docker run --rm node:24-slim sh -c "npm install -g @sleep2agi/agent-network-dashboard@0.5.3-preview.15 && grep..." ✅
JSX copy verified at source level: at app/components/ServersDrawer.tsx — no dependency on the .next/server bundled output
v0.10.x patch density of 8 patches in one day validates that the audit-first cadence is sustainable

Release stats — v0.10.8

18 cumulative @latest publishes (v0.9.0 → v0.10.8): 0 split-brain / 0 rollback / 0 retry
2026-05-17 v0.10.x same-day ships: v0.10.1-8 = 8 ships in ~11 hours (audit-first cadence)
Vincent catch + Vincent #158 LOCKED directive closed out in the same cycle

See the v0.10.8 release notes.

v0.10.7 — codex-sdk batch path yolo flags parity (2026-05-17) ✅ stable

Version alignment (npm latest tag):

@sleep2agi/agent-network@2.2.6 ← bumped (codex-sdk batch path yolo parity)
@sleep2agi/agent-node@2.4.2 (unchanged)
@sleep2agi/commhub-server@0.8.2 (unchanged)
@sleep2agi/agent-network-dashboard@0.5.2 (unchanged)

Fix — #156 codex-sdk batch path yolo flags parity

Vincent catch: "fast 也要开新一下默认 / codex 默认 fast 啊" (codex needs fast mode by default in the batch path too).

Pre-fix vs Post-fix matrix:

Path	Pre-fix	Post-fix
`anet node create --runtime codex-sdk`	✅ 4/4 yolo flags (#149 v0.10.3 ship)	✅ 4/4 yolo flags (unchanged)
`anet create --batch --runtime codex-sdk`	❌ only 1/4 (`dangerouslySkipPermissions` baseline only)	✅ 4/4 yolo flags (matches single-node)
`anet [...] --runtime codex-sdk --no-yolo`	(flag did not exist)	✅ 1/4 baseline only (opt-out for CI/scripted)

Implementation — clean helper extraction at bin/cli.ts:125-131:

function codexSdkYoloFlags(noYolo?: boolean): Record<string, string | boolean> {
  if (noYolo) return {};
  return {
    approvalPolicy: "never",
    sandboxMode: "danger-full-access",
    skipGitRepoCheck: true,
  };
}

Single source-of-truth helper: both the single-node path (cli.ts:1146) and the batch path (cli.ts:6223) call the same function — eliminates the v0.10.6 1/4-vs-4/4 drift. dangerouslySkipPermissions: true is the baseline, set at each call site, 1+3=4 yolo flags total.

User impact

Batch + codex-sdk users: previously, batch-wizard-created codex agents would block on tool-approval popups / sandbox / git checks, losing the yolo autonomous posture → now all 4 flags are set, autonomous behavior matches single-node.
Single-node users: unaffected (path unchanged).
CI / scripted users: the new --no-yolo flag provides an explicit safe-mode opt-out.
Non-codex-sdk runtimes (claude / sdk): completely unaffected (helper is gated on runtime === "codex-sdk").

Quality gates + lessons

Source-grep verify: grep -n against bin/cli.ts HEAD across 5 sites all PASS (helper + 2 call sites + wiring + field).
Docker container smoke: Cell A anet login setup failed (test infra blocker, not a fix bug) → Gate 2 source-grep evidence accepted as substitute, per v0.10.6 precedent.
Docker smoke gets its token via direct curl API calls (new): Docker smoke entry scripts must use curl direct API calls to /api/auth/register + /api/auth/login to obtain a token; do not use interactive anet login (it stalls in non-TTY containers — the hub login call blocks).

Release stats — v0.10.7

17 cumulative @latest publishes (v0.9.0 → v0.10.7): 0 split-brain / 0 rollback / 0 retry
2026-05-17 v0.10.x same-day ships: v0.10.1-7 = 7 ships in ~10 hours (audit-first cadence)
10 user-feedback items all closed-loop: including v0.10.7 #156

See the v0.10.7 release notes.

v0.10.6 — `anet upgrade` Option B detached spawn + `anet create --batch` wizard silent-exit fix (2026-05-17) ✅ stable

Version alignment (npm latest tag):

@sleep2agi/agent-network@2.2.5 ← bumped (CLI upgrade + wizard fixes)
@sleep2agi/agent-node@2.4.2 (unchanged, v0.10.3)
@sleep2agi/commhub-server@0.8.2 (unchanged)
@sleep2agi/agent-network-dashboard@0.5.2 (unchanged, v0.10.4)

Chicken-and-egg upgrade note — one-time manual install required

v0.10.4's #151 Option A only updated the verbiage (anet upgrade now shows "⚠️ NEEDS MANUAL UPGRADE"), but the chicken-and-egg deadlock wasn't fixed — your current 2.2.2 / 2.2.3 / 2.2.4 binary still falls back to the old "skipped (would replace running CLI)" behavior (that logic is frozen on the npm tarball).

bash

npm install -g @sleep2agi/agent-network@2.2.5    # one-time manual install
anet --version                                    # expect v2.2.5

From the next release onward (e.g. 2.2.6+), anet upgrade will auto detached-spawn the install for you — no more manual install.

Fixes

#154 anet upgrade Option B detached spawn enabled by default (Vincent catch): users saw anet (self): skipped (would replace the running CLI). + [anet] Done. and assumed success, but the anet binary never actually upgraded (chicken-and-egg deadlock — a Node process can't in-place replace its own binary). bin/cli.ts:3873-3874 now does spawn(forkScript, [], { stdio: "inherit", detached: true }) + child.unref() + main process process.exit(0); the detached child runs npm install in the background. The new version takes effect 1-2 min later — no --self flag needed.
#155 anet create --batch wizard silent-exit fix (Vincent catch): after the workdir-mode select(), process.stdin state changes and the readline-based ask() helper returns at EOF immediately → the entire wizard silently exits at the Node prefix prompt (same root cause as #137 in v0.9.2 preview.5 anet create regression, recurring in a different code path). Fix: migrate all post-select prompts to inquirer.input() so stdin handling stays uniform with the preceding select; the catch fallback retains the legacy ask() for non-TTY / no-inquirer environments.

Quality gates + lessons

Docker smoke gate is never skipped (Vincent): the v0.10.4 emergency trust path is SUSPENDED — the Docker smoke gate is never bypassed again.
All test nodes run in Docker (Vincent): red-line — all test nodes go in Docker, must never connect to a host hub.
dist is an obfuscated bundle — verify against source (new): dist/bin/cli.js is esbuild bundled + obfuscated (rotating string table, mangled identifiers, encoded string literals) — static grep on dist is useless. Code-path verification must grep bin/cli.ts source (HEAD = the preview build source).

Release stats — v0.10.6

16 cumulative @latest publishes (v0.9.0 → v0.10.6): 0 split-brain / 0 rollback / 0 retry
2026-05-17 v0.10.x same-day ships: v0.10.1 + v0.10.2 + v0.10.3 + v0.10.4 + v0.10.5 + v0.10.6 = 6 ships in ~9 hours (audit-first cadence)
9 user-feedback items all closed-loop: Install/Upgrade docs split + #149 / #150 / #151 / #152 / #153 / #154 / #155 + red-line SOPs

See the v0.10.6 release notes.

v0.10.5 — `anet create --batch` wizard double-fix (2026-05-17) ✅ stable

Version alignment (npm latest tag):

@sleep2agi/agent-network@2.2.4 ← bumped (CLI wizard fixes)
@sleep2agi/agent-node@2.4.2 (unchanged, shipped in v0.10.3)
@sleep2agi/commhub-server@0.8.2 (unchanged)
@sleep2agi/agent-network-dashboard@0.5.2 (unchanged, shipped in v0.10.4)

Fixes

#152 anet create --batch wizard now prompts for workdir mode (Vincent push): the --workdir-mode <shared|separate> flag has shipped since #55, but the interactive wizard never prompted — users had to know the flag name to change the default separate. createBatchWizardCommand now adds an inquirer select (separate default / shared co-cwd), TTY-aware (flag set / non-TTY both fall back to separate with an INFO hint). agent-node / server / dashboard untouched — purely CLI wizard UX.
#153 codex-sdk / claude-code-cli runtime selection no longer falsely prompts for ANTHROPIC_AUTH_TOKEN (Vincent push): the runtime-first wizard (#133) called selectVendorAndModel() even when codex-sdk / claude-code-cli was picked (only claude-agent-sdk needs an API key). The wizard now skips the API key prompt for those runtimes and prints a one-line codex auth login / claude auth login hint instead.

See the v0.10.5 release notes.

v0.10.4 — `anet upgrade` UX warning + Dashboard orphan-band layout (2026-05-17) ✅ stable

Version alignment (npm latest tag):

@sleep2agi/agent-network@2.2.3 ← bumped (#151 anet upgrade UX)
@sleep2agi/agent-network-dashboard@0.5.2 ← bumped (#150 orphan-band layout)
@sleep2agi/agent-node@2.4.2 (unchanged, shipped in v0.10.3)
@sleep2agi/commhub-server@0.8.2 (unchanged)

Fixes

#151 anet upgrade self-skip warning is now explicit (Vincent push): the prior anet (self) — self-skip row didn't spell out why or how; users walked the plan and didn't realize their own version wasn't bumped. Now adds an explicit warning + guides toward the --self flag.
#150 Dashboard topology orphan nodes now collected into an "Others" cluster box (Vincent push): previously, orphan nodes (no prefix group) were scattered across the canvas and hard to find; they now collect into an "Others" cluster box rendered alongside the other groups.

Vincent emergency trust path

v0.10.4 was shipped via Vincent's emergency trust path, skipping the test-lead Docker smoke gate (the no-testing-on-prod rule was not waived, but Vincent took the lead-scope trust path). Docker smoke is still the release-gate playbook standard checkpoint, unchanged.

See the v0.10.4 release notes.

v0.10.3 — codex-sdk default model now gpt-5.5 + yolo flags visible in config (2026-05-17) ✅ stable

Version alignment (npm latest tag):

@sleep2agi/agent-network@2.2.2 ← bumped (cli.ts vendor preset)
@sleep2agi/agent-node@2.4.2 ← bumped (codex-sdk runtime + flags)
@sleep2agi/commhub-server@0.8.2 (unchanged)
@sleep2agi/agent-network-dashboard@0.5.1 (unchanged)

Fixes

#149 codex-sdk default model fix + yolo flags written into config (Vincent catch):
- cli.ts codex vendor preset default model placeholder gpt-5.4 → real gpt-5.5
- codex-sdk runtime gains yolo: true flags (same concept as the Claude Code preset's dangerouslySkipPermissions + teammateMode) — skips the permission-prompt for multi-agent batch runs
- Flags are persisted in config.json rather than as an ephemeral runtime arg, so users can inspect / edit them

See the v0.10.3 release notes.

v0.10.2 — Hero A disk telemetry + Hero D topology label UX (2026-05-17) ✅ stable

Version alignment (npm latest tag):

@sleep2agi/agent-network@2.2.1 (unchanged — PINNED_SERVER_VERSION stays at 0.8.2)
@sleep2agi/agent-node@2.4.1 ← 2.4.0 (Hero A disk telemetry, additive; commit 50d25b2)
@sleep2agi/commhub-server@0.8.2 (unchanged)
@sleep2agi/agent-network-dashboard@0.5.1 ← 0.5.0 (Hero D topology prefix-label Option C + disk render + 100+ rounds of polish)

Hero A — agent-node disk telemetry (#99 per-server daemon Phase 2 host metrics, final 10%)

agent-node/src/host-telemetry.ts adds readDiskStats() via execFileSync('df', ['-k', '/']) (+33 lines):

POSIX -k standardizes KB output, so Linux and macOS share a single parse path
No shell pipe (per the recent shell-audit safety review) — execFileSync direct call
Windows / parse failure: graceful null (the dashboard renders — rather than a misleading 0)
HostTelemetry interface gains disk_total_gb / disk_used_gb / disk_avail_gb; getHostTelemetry() composes disk via toGb() on the same path as mem/cpu
Backward compat: older servers silently drop unknown keys; agents and servers upgrade independently

Wires through RFC-014 — GET /api/server/:host/health now returns disk's three fields, the 24h bucketed history includes disk_avail_min / disk_used_max, and alert_level adds disk < 1GB critical / < 5GB warn triggers (server/src/index.ts:253-258).

Test lead Docker Linux smoke 3/3 PASS (disk 299.8 GB total / 216 used / 71.5 avail, alert green, backward compat verified).

Hero D — Dashboard topology prefix-label UX, Option C (dashboard `0.5.1`)

#147 (acked 5/16) + Option C landed:

Topology node prefix labels (the node→group edge labels' distinguishability) ship with the Option C design (dashboard team design pass)
Disk telemetry hover-card rendering (disk_total_gb / disk_used_gb / disk_avail_gb wired to the GET /api/server/:host/health response)
100+ rounds of typography + corner-radius cascade polish

Dashboard team design pass + 4/4 verify (commit 7de97ee + screenshot evidence, local ship f9c83cd).

Closed issues

#99 per-server daemon Phase 2 close gate met (host metrics fully wired, Hero A disk shipped)
#147 Hero D (5/16 ack → Option C shipped)

RFC artifacts preserved (v0.12.0 scope)

v0.10.2 is a hotfix scope (no v0.11.0-series RFC ship); three RFC artifacts are preserved as v0.12.0 candidates:

RFC-013 v5 rename hot-reload (third-pass review complete)
RFC-014 v2 daemon Phase 2 host metrics (Hero A final 10% shipped in v0.10.2)
RFC-015 v2 #114 token-usage UI (first-pass REVISION complete)

Upgrade

bash

anet upgrade                                     # bumps agent-node 2.4.0 → 2.4.1 + dashboard 0.5.0 → 0.5.1
anet project restart                             # restart the project (pulls the new agent-node + dashboard)

Migration / Breaking

No breaking changes — the disk fields are additive (the HostTelemetry interface gains three fields, the server schema silently drops unknown keys for backward compat). Agent and server upgrade independently; for agents that don't emit the field, SQL stays NULL and the dashboard renders — rather than a misleading 0.

Release flow follows the v0.9.0 split-brain lessons #126 two-phase publish SOP.

v0.10.1 — Hotfix: PINNED_SERVER_VERSION chain-bump after the v0.10.0 ship (2026-05-17) ✅ stable

Version alignment (npm latest tag):

@sleep2agi/agent-network@2.2.1
@sleep2agi/agent-node@2.4.0 (unchanged)
@sleep2agi/commhub-server@0.8.2 (unchanged)
@sleep2agi/agent-network-dashboard@0.5.0 (unchanged)

Fix

agent-network/bin/cli.ts:61 PINNED_SERVER_VERSION was never bumped across the v0.9.x + v0.10.0 promotes — it stayed hardcoded at 0.8.0. That meant anet hub start was actually running bunx --bun @sleep2agi/commhub-server@0.8.0 (cli.ts:2589) — the old server, not the v0.10.0-shipped 0.8.2. Direct impact:

The #99 per-server daemon endpoints GET /api/server/:host/health + GET /api/server/:host/agents don't exist in 0.8.0 → 404
#142 server schema alignment for process_telemetry isn't wired in 0.8.0 → the older schema silently drops the field
Dashboard 0.5.0 §3.F server-health ring tint had no data source / §3.E hover card process_telemetry rendered all-null

A v0.10.0-announced functionality regression on the default anet hub start workflow (manually launching bunx --bun @sleep2agi/commhub-server@latest was not affected).

Fix (commit 4d24024):

diff

- const PINNED_SERVER_VERSION = "0.8.0";
+ const PINNED_SERVER_VERSION = "0.8.2";

Upgrade

bash

anet upgrade                                     # bumps agent-network 2.2.0 → 2.2.1
anet project restart                             # restart the project (pulls the new commhub-server)

Or fresh install:

bash

npm install -g @sleep2agi/agent-network@latest

Lessons

release-gate playbook now covers PINNED chain-bump — every promote-to-latest must chain-bump PINNED_*_VERSION (server pin / dashboard pin / agent-node pin), or the default path keeps running the previous ship. Consistent with #80 PINNED bump SOP, but the Hero 4 release-gate playbook hadn't covered it; the case is now added (see lesson memory in methodology).

Release flow follows the v0.9.0 split-brain lessons #126 two-phase publish SOP.

v0.10.0 — Direct Runtime + Observability Foundations (2026-05-16) ✅ stable (Phase 1, 3-package promote)

Version alignment (npm latest tag, Phase 1):

@sleep2agi/agent-network@2.2.0
@sleep2agi/agent-node@2.4.0
@sleep2agi/commhub-server@0.8.2
@sleep2agi/agent-network-dashboard@0.5.0 ✅ Phase 2 shipped

Theme: fix the root

v0.7 → v0.9.2 accumulated 11 releases; the 5-P0 ripple chain (#135-#139) surfaced a runtime architecture debt. v0.10.0 fixes the root — runtime debt (codex-sdk wrapper bypass, opt-in) + observability foundations (per-server daemon endpoints + per-agent process telemetry) + release-gate playbook. This is the groundwork for v0.11.0's 24/7 multi-vendor AI Agent society live stream. See the v0.10.0 release tracker #140.

5 features

A. #141 codex app-server stdio direct (opt-in ANET_CODEX_STDIO_DIRECT=1) The codex runtime previously went through the @openai/codex-sdk npm wrapper; the wrapper's --mcp-config HTTP transport bug is the root-cause family for the #102 hang. v0.10.0 adds a direct spawn('codex', ['app-server']) + minimal stdio JSON-RPC client (~155 LOC) path that bypasses the wrapper entirely, sidesteps that family bug, and exposes the full 67-method v2 protocol surface (thread / turn / item / realtime). v0.10.0 still defaults to the wrapper path (collecting preview feedback first); set ANET_CODEX_STDIO_DIRECT=1 to opt in. Default flip is planned for v0.11.0.

B. #99 Per-server daemon Phase 1 scaffold (monitoring-only) New server-side endpoint family (commit e575cc6):

GET /api/server/:host/health — host CPU / mem / disk / process health
GET /api/server/:host/agents — per-host agent list + telemetry history

Dashboard integration lands in Phase 2 (#119 ServersDrawer integration). The control layer (kill / restart / redeploy) is deferred to v0.11.0.

C. #142 Per-agent process telemetry agent-node now embeds process_telemetry in every commhub_report_status heartbeat: rss / cpu_pct / uptime_seconds / in_flight_count. Zero sysmon dependency, zero privilege. commhub-server schema is aligned on the wire (commit 209cac7); the dashboard hover-card rendering ships in Phase 2 (#119 sibling). Sibling to #119 host fields (host step 1 ✅ earlier; agent step 2 ships in this release).

D. Dashboard network/node front-end surface upgrade (Hero 3 — 8/8 surfaces complete, dashboard 0.5.0)

§3.A prefix-group fix (Vincent #1 catch)
§3.B sweep retire (legacy sweep path merged into grid)
§3.C recent-panel hide
§3.D grid default view
§3.E hover detail card
§3.F server-health ring tint (wired to #99 endpoint)
§3.G fullscreen mode
§3.I canvas brand mark
(§3.H dropped per RFC Q2 review)

Ships alongside 19+ rounds of typography + corner-radius cascade polish (four typography families + systematic corner-radius cascade rework). Dashboard 0.5.0 is now on the npm latest tag, landing with this v0.10.0 Phase 2 docs sync.

E. Lightweight pre-release-gate playbook (release-gate Phase 1+2)docs/tests/release-gate-playbook.md — maintained by the test lead. Covers hub / dashboard / login / node lifecycle / runtime smoke / vendor verify. v0.10.0 is the first release to fully exercise this playbook; future latest promotes gate on it.

Breaking changes / Migration

codex runtime default behavior unchanged: still goes through the @openai/codex-sdk wrapper; set ANET_CODEX_STDIO_DIRECT=1 to switch to the direct stdio path. The preview-chain ANET_CODEX_LEGACY_SDK=1 fallback flag is renamed to the more direct opt-in ANET_CODEX_STDIO_DIRECT=1 for latest — same semantics, clearer name.
agent-node commhub_report_status payload adds process_telemetry sub-object: commhub-server 0.8.2 schema is aligned; older server versions (≤ 0.8.1) silently ignore the unknown field and won't fail.
/api/server/:host/health + /api/server/:host/agents are new endpoints: rate-limit / auth behavior matches the existing /api/servers (admin or self-network member); no existing client is broken.

Known follow-ups

~~Phase 2 dashboard 0.5.0 promote~~ ✅ shipped (dashboard 0.5.0 landed with this v0.10.0 Phase 2 docs sync — 8/8 Hero 3 surfaces complete + 19+ rounds of polish)
Close evidence for #102 / #103 hang — Phase 1.5 in the preview chain already plans the regression replay against the new stdio path; we'll close them once the latest opt-in path passes the regression.
Sessions NETWORK column display bug (Vincent catch) — deferred to a v0.10.x patch or v0.11.0.
#117 anet project up detached-tmux follow-up (macOS bun setRawMode already fixed by #136, but detached-mode follow-up still pending) — to be done with the v0.11.0 control layer.

Release flow follows the v0.9.0 split-brain lessons #126 two-phase publish SOP: publish each tarball with --tag preview first, curl-verify HTTP 200, then npm dist-tag add @<v> latest. Phase 1 three-package (agent-network / agent-node / commhub-server) clean-semver promote is complete; dashboard Phase 2 is pending §3.D/F/G.

v0.9.2 — Patch: Auth fast-fail + fan-out retry + wizard redo + #122 default-tmux reverted (2026-05-16) ✅ stable

Version sync (npm latest tag):

@sleep2agi/agent-network@2.1.15
@sleep2agi/agent-node@2.3.10
@sleep2agi/commhub-server@0.8.1 (unchanged)
@sleep2agi/agent-network-dashboard@0.4.6 (unchanged)

The release flow follows the v0.9.0 split-brain lessons (#126) two-phase publish SOP (first --tag preview to upload + curl-verify the tarball returns HTTP 200, then npm dist-tag add @<v> latest), which avoids the npm publish --tag latest direct path's split-brain in the CDN async window. The version numbers themselves are clean semver (no -preview.N suffix — easier for anet upgrade comparison + npm range resolution).

5 P0 fix chain

1. #129 Vendor API auth fast-fail + vendor-specific URL hint (9840cf3) Background: when an intern key expired, agent-node previously silently waited 120s then printed a generic "claude-agent-sdk 调用超时" error. Now the isAuthError(msg) heuristic covers the Anthropic standard (401/403, invalid_api_key, authentication_error) plus intern's A02xx family (user_token_expired) plus generic OpenAI-compat 401 envelopes (unauthorized / expired_token); on hit it short-circuits the retry loop (retrying with the same bad key just wastes the backoff window) and prints a vendor-specific URL hint based on ANTHROPIC_BASE_URL host (intern → chat.intern-ai.org.cn, minimax → platform.minimaxi.com, anthropic → console.anthropic.com, otherwise generic). The clear error returns in under 5 seconds instead of the 15 minutes pre-v0.9.1 took. Full write-up: troubleshooting → Vendor API auth failure.

2. #132 Tier 1 Fan-out timeout + retry-with-backoff (same commit 9840cf3) The SDK concurrency investigation Phase 3: in a 30-agent papercope fan-out demo, intern API per-request latency stretched to 17-37s (10-20× the 1.57s single-agent baseline), so the old 120s timeout fired mid-stream and 25 / 30 sub-agents silently failed. Two changes:

CLAUDE_TIMEOUT_MS default 120000 → 300000 (300s, leaves room for the intern queue to drain)
New CLAUDE_MAX_RETRIES env var, default 2 (3 attempts total); on transient errors / timeouts, backoff 4s, 8s + 0-1s jitter (jitter spreads herd retries across the recovering vendor queue); auth-class errors do not retry (handled by #129 fast-fail above)

Set CLAUDE_MAX_RETRIES=0 to revert to the v0.9.1 no-retry behavior.

3. #133 anet node create runtime-first wizard (29fd290) Vincent catch: the old vendor-first selector enumerated only claude-agent-sdk vendors (intern / MiniMax / Claude / GLM / ...), leaving users who want claude-code-cli (Anthropic Max plan + local claude CLI login) or codex-sdk (OpenAI codex auth login) implicitly stuck — they had to know to pass --runtime codex-sdk to skip the vendor picker. New flow:

selectRuntime() 3-way picker: claude-agent-sdk / claude-code-cli / codex-sdk
claude-agent-sdk → continues into selectVendorAndModel() (existing flow)
claude-code-cli → prints claude auth login hint, skips vendor
codex-sdk → prints codex auth login hint, skips vendor

Backward-compatible: explicit --runtime <X> still skips the picker; demo / batch / scripted callers that already inject credentials via --env are unaffected.

4. #136 Revert v0.9.0 #122 default-detached-tmux (a3a3fd4) Background: detached tmux + bun claude-code-cli calling setRawMode triggers errno 5 (EIO) on macOS — the detached child's stdio is not a real PTY. Reverts the brief v0.9.0 four-condition wrap matrix:

anet node start <alias> → foreground by default (fixes the macOS bug)
anet node start <alias> --tmux → tmux new -As <alias> in attached mode (keeps the PTY chain intact, no setRawMode bug)

Removes the --foreground / --no-tmux / --attach flags (foreground is the default; --tmux already attaches). anet project up's internal startNodeTmuxSession path still uses detached (which can re-trigger setRawMode on macOS bun — follow-up tracking).

5. Wizard silent-exit fix chain #135 / #138 / #139

#135 Node v24 top-level-await warning (fa08eb4 v3): wrap dispatch in an async main() to fix the root cause
#138 @inquirer/prompts + readline ask() stdin mismatch (b8c5885 preview.5 + 596cfe9 preview.6): the wizard silently exiting after select() is fixed + launchAgent now awaits its child before the parent exits
#139 dispatch — add await to 5 async commands (15cf6de preview.7)

Pairs with #133's runtime-first wizard to keep the interactive wizard stable on Node v24.

6 SOP updates

Two-phase publish SOP continues (v0.9.0 split-brain lessons #126) — avoid the npm --tag latest direct path
macOS PTY edge case enters the release smoke checklist (detached tmux + setRawMode)
Fan-out / high-concurrency timeout tuning is documented: troubleshooting Vendor API timeout section + agent-node.md env table
Runtime-first wizard split: anet create vs --batch — the batch wizard is still vendor-first (it uses selectVendorAndModel() but does not go through selectRuntime()), and the docs now disambiguate explicitly
CLAUDE_MAX_RETRIES=0 opt-out — keeps a path back to v0.9.1 no-retry behavior for debugging / legacy scripts
Vendor-specific remediation-hint routing is now a design principle — every newly verified vendor must ship a remediation URL hint

Breaking changes / Migration

⚠ anet node start default behavior swung again (v0.9.0 → v0.9.1 brief detached → v0.9.2 reverted to foreground): the --foreground / --no-tmux flags scripts/CI added in v0.9.1 are removed in v0.9.2 (foreground is the default, no flag needed); for tmux, opt in with --tmux (attached mode)
⚠ anet node create interactive flow changed (vendor-first → runtime-first): scripted --runtime <X> still skips the picker, so no backward-compat issue; the interactive UX improves (all 3 runtimes are first-class)
⚠ CLAUDE_TIMEOUT_MS default 120 → 300: scripts that already explicitly override to a shorter timeout are unaffected; the default is more tolerant of fan-out scenarios
✅ New CLAUDE_MAX_RETRIES env var: default 2 (3 attempts total). Set 0 to revert to v0.9.1's no-retry behavior

v0.9.1 — Patch: #130 intern tool-calling hotfix promoted (2026-05-15) ✅ stable

Version sync (npm latest tag):

@sleep2agi/agent-network@2.1.14 (version-only bump, no source changes; lets anet upgrade show a v0.9.1 line that aligns with agent-node 2.3.9 + dashboard)
@sleep2agi/agent-node@2.3.9 (promotes the #130 hotfix to latest)
@sleep2agi/commhub-server@0.8.1 (unchanged)
@sleep2agi/agent-network-dashboard@0.4.6 (unchanged)

Fixes

#130 intern-s2-preview tool calling now works (commit 4cd0024 + two-phase publish promote) — in v0.9.0 the intern-s2-preview endpoint, on the Anthropic protocol with tool_choice: "auto", defaulted to verbose Thinking Process text and did not emit tool_use content blocks. Forcing tool_choice was rejected with -20077. Hotfix: when ANTHROPIC_BASE_URL matches intern-ai.org.cn / chat.intern-ai, a short system-prompt bias is prepended that nudges the model to emit tool_use directly. Direct-curl A/B verified: stop_reason: max_tokens → tool_use, output_tokens: 1024 → 122. See Vendor Adapters for the full mechanism, the 5 side effects, and the opt-out path.

Known gaps (non-blocking)

The vendor adapter relies on a URL regex to detect the vendor — self-hosted lmdeploy / proxied intern endpoints / aggregator routes will not trigger the bias and need a manual --prompt. See Vendor Adapters — Side effects.
The --no-vendor-bias flag is not yet implemented (P1 polish gap, planned alongside a bias_active info-display follow-up).

Release process

Per the v0.9.0 split-brain lessons (issue #126), uses the two-phase publish path:

1. version bump → preview.N+1
2. npm publish --tag preview      → tarball uploaded
3. curl tarball URL                → HTTP 200 confirmed
4. npm dist-tag add @<pkg>@<v> latest
5. npm view dist-tags.latest       → verifies "<v>"

This avoids the direct npm publish --tag latest path that splits during the CDN async window, where the latest tag points at a version whose tarball is not yet served everywhere.

v0.9.0 — Recovery & Observability (2026-05-15) ✅ stable

@sleep2agi/agent-network@2.1.13
@sleep2agi/agent-node@2.3.8
@sleep2agi/commhub-server@0.8.1
@sleep2agi/agent-network-dashboard@0.4.6

🎯 Theme: Recovery & Observability

The full zero-keystroke recovery loop for 22-node reboots + transparent default toolset behavior + server-level aggregate observability.

New features — Recovery chain

anet project up / restart / down (issue #117) — cwd-wide node orchestration; scans .anet/nodes/ and starts / restarts / stops every node. Shared options: --stagger <seconds> (default 3) / --only a,b,c / --exclude x,y. down caps the hub-offline notify at a 2-second race so a crashed-hub teardown for 22 nodes doesn't deadlock.
anet node create --resume <id> / --resume-latest (issue #115) — bind an existing Claude session at node-create time; TTY mode launches an interactive picker listing ~/.claude/projects/<cwd>/*.jsonl (age / size / 60-char first-line preview). anet session ls and the picker share the same listClaudeSessions() helper.
Zero-keystroke recovery (#115) — anet node start injects CLAUDE_CODE_RESUME_THRESHOLD_MINUTES=999999999 into the claude spawn env, skipping Claude Code's default 70-minute session-age threshold for the "Resume from summary / full / Don't ask again" interactive prompt. Per-spawn, does not pollute ~/.claude/settings.json, respects an explicit user override. Resume restores the full session as-is (no per-invocation flag forces a compact summary; restart-recovery is safer without surprise compaction).
anet node start auto-wraps into detached tmux (issue #122, ⚠ reverted in v0.9.2, see #136) — v0.9.0 briefly introduced a 4-condition wrap matrix (TTY + $TMUX not set + tmux installed + no same-name session) with --foreground / --no-tmux / --attach flags and a two-layer recursion guard. v0.9.2 reverts: macOS bun triggered setRawMode errno 5 under detached tmux (the detached child's stdio isn't a real PTY), so the new design is foreground-by-default + --tmux opt-in attached mode (keeps the PTY chain intact). anet node stop still kills the tmux session before SIGTERM.
anet upgrade overhaul — 4 packages, dual-channel, dry-run, self (issue #88) — covers anet self / agent-node / commhub-server / dashboard. Channel auto-detected (prerelease tag → preview, else latest); --channel overrides. --dry-run prints the plan only. --self is an opt-in detached spawn (default prints the manual command to avoid replacing the running CLI mid-upgrade). Plan rows carry action badges: upgrade / up-to-date / lazy via npx skip / self skip / lookup failed. The commhub-server row always shows PINNED_SERVER_VERSION = 0.8.0 as a reminder that anet hub start runs the pinned version regardless of what's globally installed.
anet.sh install / upgrade scripts sync (issue #123) — the anet.sh one-shot scripts now match npm's dual-channel layout, with the Node 22.13 engine check baked in.

New features — Runtime default-behavior transparency

claude-agent-sdk default = Claude Code preset (issue #101 Option B) — root-cause fix: with no tools field in config.json, agent-node was passing the SDK options.tools = undefined, giving the agent zero built-in tools and producing hallucinated "network restricted" responses. Now agent-node falls back to the SDK { type: 'preset', preset: 'claude_code' } sentinel — every agent gets WebFetch / WebSearch / Bash / Read / Write / Edit / Glob / Grep / Task / NotebookEdit by default. --tools "all" routes to the same preset (replaces the old hardcoded 8-tool list as the single source of truth).
Behavior-disclosure banner (#101, per Vincent) — anet node create prints a banner with the built-in tools + MCP tools + dangerouslySkipPermissions=true warning + restrict-tools / disable-auto-skip / inspect-current-set hints. anet info <alias> displays tools: + flags: lines for ad-hoc audits.
anet ls -v / --verbose (companion to #101) — prints a second line per node with tools=... permGate=on/off.

New features — Security hardening

Vendor token envRef mode (issue #125, v0.9.0 P0 gate #2) — the config.json env map now accepts a tagged union: string (legacy, still works, with a one-shot deprecation banner) or { "_envRef": "VAR_NAME" } (recommended — the secret stays in process.env and never touches disk). agent-node refuses to start (FATAL with remediation hint) if a referenced env var is unset — no more silent broken startup.
anet node create auto-rewrites secrets — saveCreatedNode runs rewritePlainSecretsToEnvRef() before the first write. The detection heuristic — key suffix /_TOKEN|_KEY|_SECRET|AUTH$/ or value prefix /sk-|utok_|ntok_|atok_|ak-|gsk_|key-|Bearer/ — flips matching values to envRef, drops the original into the current process.env (so the immediate spawn works), and prints export NAME='value' lines for the user to persist in ~/.bashrc.
anet node migrate-token-to-envref <alias> — new command for migrating existing nodes in place. Writes config.json.bak-<ts>, rewrites + prints export lines; idempotent (non-secret and already-migrated values are left alone).
anet doctor enumerates plain-secret nodes — passive scan + migrate suggestion (no --fix; per-node opt-in).

New features — Observability

GET /api/servers REST endpoint (issue #119, server commit 11a3018) — aggregates agents by hostname + ip and returns live host telemetry, used by the dashboard's "Servers" sidebar. Returns a bare JSON array (not the {ok, ...} wrapper). Marks 10-min-stale sessions offline before aggregating; network-scoped via addNetworkScope. Fields: hostname / ip / agent_count / cpu_load_1min / cpu_cores / mem_avail_gb / mem_used_gb / last_seen.
agent-node host telemetry (#119 step 1, commit 5364931) — each report_status call now carries host fields. On Linux: /proc/loadavg + /proc/meminfo MemAvailable first. On macOS/Windows: falls back to os.loadavg() / os.totalmem() / os.freemem(). Windows [0,0,0] is actively coerced to null. A 10-second cache prevents burst reports.
Dashboard ServersDrawer (#119 step 3) — UI sidebar shows aggregated agent counts per physical machine alongside live CPU / RAM bars.
Dashboard topology redo + 38 rounds of polish (issues #112 + #116) — grid + ring dual views; mount fade-in; hover ring focus; click ripple; label scaling; arrow tiers; offline dim; group-box hover; minimap; cwd tooltip; and 9+ further rounds of interaction polish.

Documentation

GitHub README front-page overhaul (issue #118, commit 2dd646d) — Hero / Quick start / Demo / CTA promoted to the top; anet vs LangGraph/AutoGen/CrewAI 5×4 comparison table; trust signals (4 new badges + Star History chart); mermaid architecture diagram + node onboarding flow; ZH + EN parity.
docs-site catch-up sweep (issue #124) — bulk-syncs every new feature shipped today to anet.sh: cli.md / upgrade.md / security.md / rest.md / CHANGELOG.md, all ZH + EN.

Breaking changes / Migration

⚠ anet node start default changed (v0.9.0 only, reverted in v0.9.2) — v0.9.0 briefly auto-wrapped into a detached tmux session (v0.8 ran foreground), but v0.9.2 via #136 reverts to foreground default (macOS bun setRawMode bug). Scripts/CI on v0.9.1 still need --foreground / --no-tmux; v0.9.2+ is foreground by default with no flag needed. For tmux, opt in with --tmux (attached mode).
⚠ claude-agent-sdk node default toolset changed — from empty to the full Claude Code preset (Bash / WebFetch / Write / …). Existing nodes with an explicit tools allowlist keep their old behavior; for new nodes, decide whether you need to narrow with --tools Read,Glob,Grep.
⚠ Vendor secrets no longer persist plain in config.json — newly created nodes go through envRef automatically; for existing plain-secret nodes run anet node migrate-token-to-envref <alias> for a one-shot migration. The plain-string path stays compatible for now (with a deprecation banner).
✅ Preview version-number rule (per Vincent) — bump the -preview.N+1 suffix within a preview chain; do not bump the patch and reset preview.0 (avoids "backwards-looking" version numbers).

Smoke validation (before promote)

1. plain fallback           — old config with "sk-..." still starts + shows the deprecation banner
2. envRef happy path        — { _envRef: "TEST_TOKEN" } + export TEST_TOKEN=fake → node receives the correct token
3. envRef missing var FATAL — case #2 without the export → startup FATAL + remediation hint
4. anet doctor scan         — mix of plain + envRef nodes → only plain ones surface under the warning
5. migrate idempotent       — plain → migrate → second run is a no-op + `.bak-<ts>` exists + export lines printed
6. anet node create auto    — `--env ANTHROPIC_AUTH_TOKEN=sk-fake` → config.json contains envRef, not literal sk-fake

See issue #125 for full repro steps.

2026-05-14 — v0.8.3 stable release batch primitive + multi-demo + P0/UX fixes ✅ stable

Version sync (npm latest tag):

@sleep2agi/agent-network@2.1.9
@sleep2agi/agent-node@2.3.1
@sleep2agi/commhub-server@0.8.0 (unchanged)
@sleep2agi/agent-network-dashboard@0.4.5

Note: agent-network 2.1.8 was skipped (an earlier stale build occupied the version); the stable release is 2.1.9.

New features

anet create --batch batch agent primitive (issue #55) — spin up N identity-bearing agents in one line, --prefix auto-numbering, per-node working dir + config + tmux session; anet batch <verb> for lifecycle management (list/stop/cleanup/start/restart).
anet demo sci-team (issue #51) — research-squad demo: 1 leader + N-1 workers with active fan-out collaboration.
anet demo pr-review (issue #41) — 4-agent PR review room demo.
anet login first-time login guidance (issue #58) — on auth failure, points to register / default account / hub admin reset-user.
claude-agent-sdk model dropdown verified vendor presets (issue #48) — MiniMax + Intern (书生).
SDK upgrades — codex-sdk / claude-agent-sdk / inquirer dependency bumps.

Fixes

batch node identity injection (issue #93, P0) — batch-created nodes previously didn't know their own alias; now per-node identity prefix is injected.
anet hub dashboard npx cache self-heal (issue #89, P0) — auto-cleans stale staging dirs before spawn.
anet hub dashboard release channel matching (issue #61) — dashboard version now dynamically matches the anet channel.
anet init token prompt UX + session count (issue #56).
shell removed from process spawn (issue #36) — eliminates command injection surface.
claude-agent-sdk env injection + timeout guard (issue #98, partial fix) — config.json env block now injected on the --config startup path; claude calls get a wall-clock timeout guard so hangs surface as visible timeout errors.
PINNED commhub-server → 0.8.0 stable.

Package change details

agent-network 2.1.7 → 2.1.9 (13 preview iterations accumulated)
agent-node 2.3.0 → 2.3.1 — claude-agent-sdk / codex-sdk dependency bumps + #98 fixes
commhub-server 0.8.0 (unchanged)
agent-network-dashboard 0.4.2 → 0.4.5 — tri-ring layout / alias avatars / fullscreen zoom / prefix grouping / Intern avatars / label-overlap fix / trial badge removal

2026-05-12 — v0.8.2 stable release telegram channel + claude-code-cli session resume ✅ stable

Version sync (npm latest tag):

@sleep2agi/agent-network@2.1.7
@sleep2agi/commhub-server@0.8.0 (unchanged)
@sleep2agi/agent-node@2.3.0 (unchanged)

Related: issue #13 (closed) · issue #14 · commit 143b2a1 (release: 2.1.7 stable) · commit f1e3d9c (fix(cli): bind claude code sessions on first start)

New features

anet channel add telegram one-shot bind — attaches a Telegram bot token + allow-user to an existing node, auto-generates the channels/telegram config (see cases/telegram-squad for details).

Fixes

claude-code-cli runtime now pre-generates a Claude session UUID when creating a node.
First start binds a fixed session with claude --session-id <uuid>; once ~/.claude/projects/<cwd>/<uuid>.jsonl exists on the local machine, subsequent starts switch to claude --resume <uuid> to continue the same conversation — anet node start no longer accidentally opens a new chat.
anet node start --new-session generates and saves a fresh session UUID.

2026-05-11 — v0.8.1 patch Dashboard SSE-online global fix ✅ stable

Version sync (npm latest tag, git tag v0.8.1):

@sleep2agi/commhub-server@0.8.0 (unchanged)
@sleep2agi/agent-network@2.1.5
@sleep2agi/agent-network-dashboard@0.4.2
@sleep2agi/agent-node@2.3.0 (unchanged)

Fixes

Dashboard /nodes, /admin, and /api/hub/session all showed every agent as offline because the SSE key in server v0.7+ became network_id:alias. The 0.4.1 fix missed these three; 0.4.2 adds alias-fallback to all global SSE lookups.
CLI bumps PINNED_DASHBOARD_VERSION to 0.4.2 so anet hub dashboard pulls the patched version.

2026-05-11 — v0.8.0 stable release 🎉 RFC-001 phase 2 landed ✅ stable

Version sync (git tag v0.8.0):

@sleep2agi/commhub-server@0.8.0
@sleep2agi/agent-network@2.1.4
@sleep2agi/agent-network-dashboard@0.4.1
@sleep2agi/agent-node@2.3.0 (unchanged)

Auth changes

COMMHUB_AUTH_TOKEN is now soft-deprecated: v0.8 keeps only /api/* read-side compat and logs a warning; v1.0 removes it entirely.
First anet hub start bootstraps a default admin account (admin / anethub quick-start default) and writes a local recovery admin utok_ to ~/.anet/server/admin-utok.json (chmod 600). Public deployments must immediately anet passwd to a strong password.
Subsequent anet hub start is idempotent: if admin-utok.json exists it skips bootstrap and no longer prompts.
Dashboard moved to browser-cookie passthrough (thin proxy mode — the full 0-token model lands later in v0.8.x).
tmux and admin endpoints now require an admin utok_.

Password management

anet passwd prompts for old / new / confirm by default; --old / --new still supported.
Successful change rotates the current device's utok_; other devices lose their utok_ automatically. Agent ntok_ are unaffected.
New: anet hub admin reset-user --username <u> — hub-host-only recovery for non-admin users, emits a password_reset_by_admin audit event.
User-chosen passwords require ≥ 8 chars + are checked against the top-1000 weak-password dictionary. The first-run bootstrap admin password is exempt (≥ 4 is accepted) since it must be rotated immediately anyway.

Doctor enhancements

anet doctor --fix now actively probes every node's ntok_ against the hub. On 401/403 it auto-reissues a fresh ntok_ from the current utok_ and patches the file in place — session_id / channels / runtime / role are all preserved. This covers the "hub DB wiped / token revoked" failure mode.

CLI / UX

anet hub start is silent auto-generate by default and no longer interrupts startup with prompts.
anet login prints ✅ and a next-step hint; double-colon prompt bug fixed.
CLI error output switched from the flat [anet] prefix to ✅ / ❌ visual markers.

Dashboard 0.4.1

Fixed Command Mesh's sse:undefined: SSE key in server v0.7+ became network_id:alias; dashboard now queries with the double-layer key, with alias-only fallback for older hubs.
Light / Mint theme solid-button polish (regressions since 0.3.4).

2026-05-10 — v2.1 stable release

Version sync (git tag v2.1.0):

@sleep2agi/agent-network@2.1.0
@sleep2agi/commhub-server@0.6.0
@sleep2agi/agent-node@2.3.0
@sleep2agi/agent-network-dashboard@0.3.0

Install

bash

npm install -g @sleep2agi/agent-network

No more @preview tag needed — latest is now the stable line.

What's new

anet doctor --fix auto-migrates legacy V2 nodes Frontline pain point: claude-code-cli runtime hit many V2-era node configs (with alias/resume/no token / dev IP for hub) that throw utok_ but SSE needs ntok_ against the V3 hub. doctor now:

Detects 6 classes of legacy config issues (renamed fields, runtime rename, stale hub, missing token, unprefixed token, missing node_id)
One-shot --fix migration; preserves the session field so chat history is not lost, re-issues ntok_

anet demo subcommand family

anet demo ls — list demos
anet demo debate — 6-agent, 9-step debate
anet demo socialmedia — 4-agent social media content factory (Xiaohongshu / Twitter / WeChat / LinkedIn)
Defaults to a standalone demo-<suffix> network, auto-cleaned afterwards — never pollutes default

Hub telemetry fixes

POST /api/task now double-writes the inbox + tasks tables (previously only the inbox, which left the Dashboard Tasks page empty and send_reply unable to find tasks)
Dispatching a task immediately UPDATEs sessions.task + updated_at so the Dashboard Overview reflects "task in flight" in real time

Dashboard theming

4 themes: Cyber (default dark) / Light / Mint / Sunset, switcher bottom-right, persisted in localStorage
Fixed the useSSE reconnect loop (the hub used to receive 1500+ admin SSE reconnects that DoSed mcp)
COMMHUB_URL fallback restored to 127.0.0.1:9200 (was a leftover dev IP)

CLI

--runtime http-api no longer falls into the Claude CLI branch
agent-node HTTP runtime now also reads ANTHROPIC_AUTH_TOKEN (previously only ANTHROPIC_API_KEY)
Demo subcommands no longer trigger 6 "select provider" interactive prompts when calling createCommand

One-shot deploy scripts

hub-only.sh rewritten: 4G swap + sudoers NOPASSWD + enable-linger + systemd autostart + AUTOSTART=1
agent-only.sh updated to match

Upgrade path

bash

# 1. Upgrade the CLI
npm install -g @sleep2agi/agent-network    # or: npm update -g

# 2. Restart the hub (so the new commhub-server takes effect)
# tmux: tmux kill-session -t hub; tmux new -d -s hub 'anet hub start'
# systemd-user: systemctl --user restart anet-hub

# 3. In every legacy project directory, run doctor --fix
cd <project-dir>
anet doctor --fix

# 4. Restart agents
kill <claude-pid> && anet resume <node-name>

See the Upgrade Guide for details.

2026-05-03 - `anet demo` subcommands and bug fixes

Version sync: anet@2.0.3-preview.4 / agent-node@2.2.0-preview.1 / dashboard@0.2.1-preview.1 / commhub-server@0.5.3-preview.0

New Features

anet demo ls - list available demos
anet demo debate - one-command 6-agent debate demo
- --topic "..." debate topic
- --key sk-cp-xxx MiniMax API key, defaulting to $MINIMAX_KEY
- --quick shortened 4-step run
- --keep keep temporary agents and network after the run
- --out path.md transcript output path
anet demo monitor - kept as the old anet demo --live alias

Fixes

anet CLI: --runtime http-api now starts through agent-node instead of falling into the Claude CLI branch.
agent-node: HTTP runtime reads ANTHROPIC_AUTH_TOKEN for MiniMax-compatible configs.
dashboard: fixed repeated SSE reconnects caused by inline onEvent callbacks.
hub-only.sh: rewritten with swap setup, sudoers NOPASSWD, linger, and optional systemd user autostart.

2026-04-30 - Parent Task Lineage + Auto-Chain Reply

commhub-server@0.5.3-preview.0

Added parent_task_id to tasks and chainReplyToParent() to keep multi-agent chains connected.
send_task accepts parent_task_id and can infer it from the caller's recent open task.
send_reply / report_completion forward results up the parent chain automatically.
agent-node injects CURRENT_TASK_ID and prompts the LLM to pass parent_task_id.

2026-04-26 - Hub Server Logs Page + V2 Lineage Foundation

commhub-server@0.5.2-preview.0 / dashboard@0.2.1-preview.0 / anet@2.0.3-preview.1

Dashboard /server-logs page for live hub stdout.
REST GET /api/server-logs for admin users.
Hub banner and /health show the published version.

2026-04-15 - V3 Stable: Multi-Network + User System + Trial License

Agent Network V3 - Multi-Network + Commercial Ready (commhub-server 0.5.x, anet 2.0.x)

Multi-network isolation for nodes, tasks, and sessions.
Username/password accounts, JWT, and the utok_ + ntok_ token system.
14-day trial licensing and Pro activation.
39 CLI commands, 17 MCP tools, and 17 REST endpoint families.
3 runtimes: claude-agent-sdk, codex-sdk, and http-api.
Audit logs, rate limiting, and PostgreSQL support through the DbAdapter.

v1.0.0-preview.25 (2026-04-11)

PostgreSQL + Adapter Architecture

New features:

PostgreSQL support: Enable via DATABASE_URL=postgres://... (SQLite remains the default)
DbAdapter interface: Unified database abstraction layer (SQLiteAdapter + PgAdapter)
SQL auto-translator: sqliteToPostgres() handles datetime->NOW, ?N->$N, AUTOINCREMENT->SERIAL
34 CLI commands: Added passwd, token (create/ls/revoke), network (info/rename/delete), demo, config, license, activate, hub start
17 REST endpoints: Added PUT /api/networks/:id, DELETE /api/networks/:id, POST /api/auth/password, token CRUD
One-click demo: bash examples/demo-one-click.sh -- 60-second automated demo
createAdapter() factory: Environment-driven database selection

Architecture improvements:

All 85+ db.query() calls migrated to adapter methods (db.get(), db.all(), db.run())
All 7 manual BEGIN/COMMIT/ROLLBACK transactions converted to db.transaction()
Zero raw database access -- all code goes through the DbAdapter interface
SQL translator handles 161 SQL fragments across 4 source files

Testing:

200 Docker E2E tests (137 core + 25 auth + 22 network + 16 config)
19 adapter-specific E2E tests
10 SQL translator unit tests

v1.0.0-preview (2026-04-10)

Agent Network V3 -- Multi-Network + Commercial Ready

New features:

Multi-network support: Create isolated networks, each with independent nodes/tasks/sessions
User system: Username + password registration/login, API token authentication
Trial licensing: 14-day free trial, license key activation for Pro
39 CLI commands: quickstart, login, register, passwd, token, network (create/ls/use/info/rename/delete), status, tasks, doctor, info, logs, demo, config, license, activate, hub start...
17 MCP tools: send_task, send_reply, retry_task, cancel_task, reassign_task, list_tasks, get_task...
17 REST endpoints: /api/auth/, /api/networks/, /api/tasks, /api/nodes, /api/stats, /api/audit-log, /api/license...
2 AI runtimes: codex-sdk (OpenAI Codex / GPT-5), claude-agent-sdk (Claude / MiniMax / OpenAI-compatible)
Audit logging: All user operations + task state changes recorded
Rate limiting: Registration 30/min, login 10/min per IP

Security:

MCP/SSE/WebSocket authentication
Server-side enforced network_id (token-bound, client cannot override)
SQL injection fix (all parameterized queries)
Network ownership checks (cross-user access returns 403)
Password hashing (SHA-256)
Localhost rate limit exemption (dev/testing)

Database (13 tables): sessions, inbox, tasks, nodes, completions, task_events, users, networks, api_tokens, audit_log, licenses, network_members, network_invites

Testing (200 regression tests):

Core E2E: 137 tests (node lifecycle, message lifecycle, auth, authorization, SSE, concurrency)
Auth suite: 25 tests (registration, login, token, profile, password, audit, rate limiting)
Network suite: 22 tests (CRUD, isolation, ownership, rename, delete, cross-user)
Config priority: 16 tests (CLI > env > project > global)
Real AI: Codex (GPT-5) + MiniMax (Anthropic API) verified
10-agent idiom chain (mixed codex + minimax)

npm packages:

@sleep2agi/agent-network (anet CLI)
@sleep2agi/agent-node (Agent runtime)
@sleep2agi/commhub-server (Communication hub)

v0.x (2026-03 ~ 2026-04-09) -- Pre-V3

Core Feature Development

CommHub Server: MCP + SSE-based communication hub
agent-node: Dual-engine runtime (Claude + Codex)
anet CLI: create / start / resume / channel and other basic commands
Dashboard: Initial version
Message types: task / reply / message / ack type differentiation
Channel plugins: Claude Code CommHub integration

Early Milestones

Version	Date	Content
v0.1	Early 2026-03	Basic CommHub + SSE
v0.3	Mid 2026-03	agent-node dual engine
v0.5	Late 2026-03	anet CLI + Channel
v0.7	Early 2026-04	Dashboard + message types
v0.9	2026-04-09	Multi-model support (MiniMax, InternLM)

Roadmap

v0.9 — Security hardening

Argon2id password hashing (currently SHA-256)
utok_ / ntok_ TTL + revoke-all
Install-script checksum verification
Dashboard full 0-token model finishing touches

v1.0 — Cleanup + public networks

Remove COMMHUB_AUTH_TOKEN compat path entirely
Token scope (full / agent / readonly) full implementation
Public / invite-hybrid networks (member application + owner approval flow)
Per-role button visibility in Dashboard

Later

~~Continued PostgreSQL adapter improvements~~ — adapter interface kept as extension point, but v0.8+ product direction is SQLite only (see docs/v3-postgresql-design.md banner)
SSO integration
Webhook callbacks
Cron-style task scheduling

Next steps

Upgrade guide — v0.7 → v0.8 behavior changes + standard steps
Architecture — how each release accumulated into the current system
GitHub Releases — per-tag release notes
RFC-001 — v0.8 ~ v1.0 master-token deprecation roadmap

Changelog ​

v0.10.13 — grok-build-acp session/prompt 300s timeout hang fix (P0 hotfix) (2026-06-08) ✅ stable ​

🌟 Highlights ​

Root cause + fix for grok-build-acp nodes hanging at session/prompt timed out after 300000ms ​

🐛 Bugs Fixed ​

📦 Install (fresh install) ​

🔄 Upgrade (existing users) ​

🙏 Credits ​

v0.10.12 — Grok-build runtime scenario enablement + 0.2.8 alpha regression verify (2026-05-30) ✅ stable ​

Highlights ​

Install / Upgrade ​

v0.10.11 — #204 grok-build-acp per-node identity isolation + #194 commhub broadcast attribution hotfix (2026-05-28) ✅ stable ​

🌟 Highlights ​

#204 — grok-build-acp per-node isolated cwd ​

#194 — commhub broadcast sender attribution hotfix ​

🐛 Bugs Fixed ​

📦 Install (fresh install) ​

🔄 Upgrade (existing users) ​

🙏 Credits ​

v0.10.10 — Xiaomi MiMo full 5-model support + envRef wizard-to-start auto-link (2026-05-27) ✅ stable ​

P0 — Xiaomi MiMo Vendor Preset complete ​

P0 — envRef wizard-to-start auto-link (#193) ​

Bug fixes ​

Known Issues ​

v0.10.9 — Dashboard image sending + CommHub attachment metadata + codex-sdk image input (2026-05-25) ✅ stable ​

P0 — Dashboard command flow can send images ​

Rollout + Smoke ​

Known Limits ​

v0.10.8 — Dashboard Servers panel UI copy fix + TopoGraph density-tier polish (2026-05-17) ✅ stable ​

Fix — #157 Dashboard Servers panel UI copy fix (Root cause #1) ​

Polish fold-in — TopoGraph density-tier (purely additive) ​

Quality gates + lessons ​

Release stats — v0.10.8 ​

v0.10.7 — codex-sdk batch path yolo flags parity (2026-05-17) ✅ stable ​

Fix — #156 codex-sdk batch path yolo flags parity ​

User impact ​

Quality gates + lessons ​

Release stats — v0.10.7 ​

v0.10.6 — anet upgrade Option B detached spawn + anet create --batch wizard silent-exit fix (2026-05-17) ✅ stable ​

Fixes ​

Quality gates + lessons ​

Release stats — v0.10.6 ​

v0.10.5 — anet create --batch wizard double-fix (2026-05-17) ✅ stable ​

Fixes ​

v0.10.4 — anet upgrade UX warning + Dashboard orphan-band layout (2026-05-17) ✅ stable ​

Fixes ​

v0.10.3 — codex-sdk default model now gpt-5.5 + yolo flags visible in config (2026-05-17) ✅ stable ​

Fixes ​

v0.10.2 — Hero A disk telemetry + Hero D topology label UX (2026-05-17) ✅ stable ​

Hero A — agent-node disk telemetry (#99 per-server daemon Phase 2 host metrics, final 10%) ​

Hero D — Dashboard topology prefix-label UX, Option C (dashboard 0.5.1) ​

Closed issues ​

RFC artifacts preserved (v0.12.0 scope) ​

Upgrade ​

Migration / Breaking ​

v0.10.1 — Hotfix: PINNED_SERVER_VERSION chain-bump after the v0.10.0 ship (2026-05-17) ✅ stable ​

Fix ​

Upgrade ​

Lessons ​

v0.10.0 — Direct Runtime + Observability Foundations (2026-05-16) ✅ stable (Phase 1, 3-package promote) ​

5 features ​

Breaking changes / Migration ​

Known follow-ups ​

v0.9.2 — Patch: Auth fast-fail + fan-out retry + wizard redo + #122 default-tmux reverted (2026-05-16) ✅ stable ​

5 P0 fix chain ​

6 SOP updates ​

Breaking changes / Migration ​

v0.9.1 — Patch: #130 intern tool-calling hotfix promoted (2026-05-15) ✅ stable ​

Fixes ​

Known gaps (non-blocking) ​

Release process ​

v0.9.0 — Recovery & Observability (2026-05-15) ✅ stable ​

🎯 Theme: Recovery & Observability ​

New features — Recovery chain ​

New features — Runtime default-behavior transparency ​

New features — Security hardening ​

New features — Observability ​

Documentation ​

Breaking changes / Migration ​

Smoke validation (before promote) ​

Changelog

v0.10.13 — `grok-build-acp` `session/prompt` 300s timeout hang fix (P0 hotfix) (2026-06-08) ✅ stable

🌟 Highlights

Root cause + fix for `grok-build-acp` nodes hanging at `session/prompt timed out after 300000ms`

🐛 Bugs Fixed

📦 Install (fresh install)

🔄 Upgrade (existing users)

🙏 Credits

v0.10.12 — Grok-build runtime scenario enablement + 0.2.8 alpha regression verify (2026-05-30) ✅ stable

Highlights

Install / Upgrade

v0.10.11 — #204 grok-build-acp per-node identity isolation + #194 commhub broadcast attribution hotfix (2026-05-28) ✅ stable

🌟 Highlights

#204 — grok-build-acp per-node isolated cwd

#194 — commhub broadcast sender attribution hotfix

🐛 Bugs Fixed

📦 Install (fresh install)

🔄 Upgrade (existing users)

🙏 Credits

v0.10.10 — Xiaomi MiMo full 5-model support + envRef wizard-to-start auto-link (2026-05-27) ✅ stable

P0 — Xiaomi MiMo Vendor Preset complete

P0 — envRef wizard-to-start auto-link (#193)

Bug fixes

Known Issues

v0.10.9 — Dashboard image sending + CommHub attachment metadata + codex-sdk image input (2026-05-25) ✅ stable

P0 — Dashboard command flow can send images

Rollout + Smoke

Known Limits

v0.10.8 — Dashboard Servers panel UI copy fix + TopoGraph density-tier polish (2026-05-17) ✅ stable

Fix — #157 Dashboard Servers panel UI copy fix (Root cause #1)

Polish fold-in — TopoGraph density-tier (purely additive)

Quality gates + lessons

Release stats — v0.10.8

v0.10.7 — codex-sdk batch path yolo flags parity (2026-05-17) ✅ stable

Fix — #156 codex-sdk batch path yolo flags parity

User impact

Quality gates + lessons

Release stats — v0.10.7

v0.10.6 — `anet upgrade` Option B detached spawn + `anet create --batch` wizard silent-exit fix (2026-05-17) ✅ stable

Fixes

Quality gates + lessons

Release stats — v0.10.6

v0.10.5 — `anet create --batch` wizard double-fix (2026-05-17) ✅ stable

Fixes

v0.10.4 — `anet upgrade` UX warning + Dashboard orphan-band layout (2026-05-17) ✅ stable

Fixes

v0.10.3 — codex-sdk default model now gpt-5.5 + yolo flags visible in config (2026-05-17) ✅ stable

Fixes

v0.10.2 — Hero A disk telemetry + Hero D topology label UX (2026-05-17) ✅ stable

Hero A — agent-node disk telemetry (#99 per-server daemon Phase 2 host metrics, final 10%)

Hero D — Dashboard topology prefix-label UX, Option C (dashboard `0.5.1`)

Closed issues

RFC artifacts preserved (v0.12.0 scope)

Upgrade

Migration / Breaking

v0.10.1 — Hotfix: PINNED_SERVER_VERSION chain-bump after the v0.10.0 ship (2026-05-17) ✅ stable

Fix

Upgrade

Lessons

v0.10.0 — Direct Runtime + Observability Foundations (2026-05-16) ✅ stable (Phase 1, 3-package promote)

5 features

Breaking changes / Migration

Known follow-ups

v0.9.2 — Patch: Auth fast-fail + fan-out retry + wizard redo + #122 default-tmux reverted (2026-05-16) ✅ stable

5 P0 fix chain

6 SOP updates

Breaking changes / Migration

v0.9.1 — Patch: #130 intern tool-calling hotfix promoted (2026-05-15) ✅ stable

Fixes

Known gaps (non-blocking)

Release process

v0.9.0 — Recovery & Observability (2026-05-15) ✅ stable

🎯 Theme: Recovery & Observability

New features — Recovery chain

New features — Runtime default-behavior transparency

New features — Security hardening

New features — Observability

Documentation

Breaking changes / Migration

Smoke validation (before promote)