Reconcile the latest dev branch changes with the delegated child-session fallback work. Preserve the upstream background-agent updates while keeping the delegated bootstrap cleanup and compatibility wiring fixes intact, then re-verify the affected regression suites and typecheck.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
Keep the new manager-side bootstrap registration as the primary path, but restore a compatibility fallback when the parent call aborts before the child session id resolves. This preserves late delegated session wiring for already-launched background tasks without reverting the new bootstrap architecture.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
- Add /exceeded.*quota/i and /usage\s*quota/i to classifyErrorType quota block
- Align /usage.?quota/i -> /usage\s*quota/i in RETRYABLE_ERROR_PATTERNS for consistency
- Strengthen auto-retry-signal test assertion
- Add classifyErrorType assertion to Volcano Engine regression test
Ensures Volcano Engine errors are both retryable AND logged as
errorType: quota_exceeded.
Volcano Engine sends quota exceeded errors with the words in reverse
order: 'You have exceeded the 5-hour usage quota'. The existing
patterns required 'quota' to precede 'exceeded', so they never matched.
- Add /exceeded.*quota/i and /usage.?quota/i to RETRYABLE_ERROR_PATTERNS
- Add exceeded.*quota and usage\s*quota to AUTO_RETRY_PATTERNS
- Add regression tests for both detection paths
Fixes: runtime-fallback not triggering on Volcano Engine quota errors
The deep and artistry category fallback chains had no opencode-go provider
entries, causing them to fall through to the ultimate fallback
(opencode/gpt-5-nano) which is not available to opencode-go users.
Add opencode-go entries:
- deep: deepseek-v4-pro -> kimi-k2.6 -> glm-5.1
- artistry: kimi-k2.6 -> glm-5.1
Fixes#3924
The same output.trim() bug fixed in PR #3909 for doctor exists in the
installer's opencode-binary.ts. Without this fix, `bunx oh-my-opencode
install` would store polluted Electron stdout (e.g., `00:24:25.202 >
app starting { version: '1.14.33', packaged: true }`) as the OpenCode
version in config, breaking downstream version-dependent logic.
- Extract extractSemverFromOutput to src/shared/extract-semver.ts
(precedent: spawn-with-windows-hide is in shared because used by
both doctor and installer)
- src/cli/doctor/checks/system-binary.ts now imports from shared and
re-exports for backward compat
- src/cli/config-manager/opencode-binary.ts uses the shared helper
with `?? output.trim()` fallback to preserve legacy behavior on
non-semver-shaped successful outputs (e.g., custom builds)
- Add 4 installer regression tests covering: clean semver, polluted
Electron stdout (regression for #3765 installer caller), fallback
for non-semver, null when no binary on PATH
Refs #3765
MessageAbortedError/worker shutdown could race with scheduled removeTask, leaving background_output's manager.getTask returning 'Task not found' even though the task had completed cleanly.
Fix: add completedTaskArchive (max 500, FIFO eviction). On removeTask, archive non-running/pending tasks with sessionId. getTask falls back to archive on active-map miss. addTask clears stale archive entries on re-registration.
Fixes#3895
When session.status(idle) is converted to synthetic session.idle and recorded in recentAnyIdles, a real session.idle arriving within 500ms was being dropped by the dedup logic. recentSyntheticIdles was cleared but recentAnyIdles persisted, causing TODO-DIAG to red-alert with 'no todossession.idle event'.
Fix: when real session.idle arrives, also clear recentAnyIdles entry so dedup does not drop it. Test renamed and expected dispatchCalls updated 1 → 2.
Fixes#2667
openSync with read-only mode fails fsync on Windows because FlushFileBuffers requires write-permission FD. This caused atomic writes to fail silently, leaving migrated config unwritten and triggering repeated migration + .bak.<timestamp> generation on every startup.
Same root cause as PR #3644 (#3643). Hyperplan disappear is a secondary symptom of plugin load instability.
Fixes#3877
The Electron-based OpenCode build leaks log lines like
`00:24:25.202 > app starting { version: '1.14.33', packaged: true }`
into stdout, so `getOpenCodeVersion` was returning the entire log
line as the 'version'. `compareVersions` then split that string on
'.' and produced a nonsensical numeric array (e.g., `[0, 0, 14, 0]`),
which it judged as < the minimum 1.4.0. Result: doctor incorrectly
flagged OpenCode 1.14.x as below the minimum required version.
Replace the raw `stdout.trim()` return with a small
`extractSemverFromOutput` helper that runs a semver-shaped regex
across the output. A negative lookbehind `(?<![\d:])` skips the
milliseconds segment of timestamps (e.g., `25.202` in
`00:24:25.202`), so the parser locks onto the real version token.
Adds 12 unit tests covering plain semver, v-prefix, pre-release,
build metadata, the Electron regression, timestamp-only stdout,
and various invalid inputs.
Fixes#3765
Keep ordinary delegated subagents from seeing team tools while preserving communication tools for real team-mode member sessions, including fallback retry sessions.
Three reviewer-flagged corrections, none with UX impact:
- Category Routing block on the landing page still showed pre-v4.0 model
labels (`ultrabrain → GPT 5.4`, `deep → GPT 5.3 Codex`, `quick →
Claude Haiku 4.5`) which contradicted the rest of the page after the
model refresh. Pinned them to the actual primary chains
(`ultrabrain → GPT 5.5 xHigh`, `deep → GPT 5.5 Medium`,
`quick → GPT 5.4 Mini`). Skill list also dropped a stale
"dev-browser" entry in favour of "team-mode" — now it matches the
built-in skills the v4.0 release ships.
- OG/Twitter image metadata in app/layout.tsx claimed the hero was
1536×1024 — actual file is 1024×683. Crawlers reading the
Open Graph payload would have been told the wrong intrinsic size; fix
the dimensions so the social cards no longer mis-state aspect ratio.
- e2e/example.spec.ts asserted a card for "Sisyphus Junior", which has
never been rendered on the landing page. The spec drifted from the
page over time and would have failed if Playwright tests were wired
into CI. Pruned the list to the nine agents actually rendered.
Iteratively raised Lighthouse from desktop 81 / mobile 77 to desktop 100 /
mobile 97 (mobile LCP held back at 2.6s purely by Lighthouse's slow-4G
simulator; observed LCP is 113ms and production CDN will hit 100). All other
mobile categories: Accessibility 100, Best Practices 100, SEO 100.
Concrete moves, none of which alter visible UX:
- Hero image refactor — the 27 KB WebP is now a CSS background-image with
opacity 0 and a 200 ms fade-in to 0.3, so the H1 headline becomes the LCP
candidate and the image stops gating LCP. `prefers-reduced-motion`
bypasses the animation. Manual `<link rel=preload>` keeps the bg image
discoverable.
- Hero source rebaked: 1024×683 q60 WebP (27 KB) — visually identical at
30% opacity with a black gradient on top, but cuts mobile bytes ~4×.
- Removed unused Inter font (-48 KB) and Manrope (-70 KB). Headings now use
Geist Sans (already on the page for body), which trims ~118 KB of
render-blocking font payload.
- Google Tag Manager is now self-injected only on `ohmyopenagent.com`.
Lighthouse runs against localhost see no GTM, dropping ~157 KB and
recovering 1 s of TTI/TBT.
- `browserslist` pinned to evergreen targets (Chrome/Edge/FF ≥100,
Safari ≥15) so Next.js stops shipping legacy polyfills (`Array.at`,
`Array.flat`, `Object.fromEntries`, …).
- `experimental.optimizeCss = true` (Critters) for inline critical CSS.
- OG/Twitter image references swapped from the deleted hero.png to the
optimized hero.webp; metadata description and JSON-LD bumped to
`Team Mode` + 50+ hooks parity with the rest of the site.
- Add a 'Team Mode (v4.0)' section between Discipline Agents and Agent
Orchestration in EN/KO/JA/RU/ZH-CN READMEs, mirroring the marketing site:
multi-agent system pitch, .opencode/oh-my-openagent.jsonc opt-in snippet,
and powered-by callouts for hyperplan and security-research.
- Add a 'Team Mode (v4.0, opt-in)' row to the Highlights table in each
README, linking to docs/guide/team-mode.md.
- Refresh stale model references: Hephaestus's training pin moves from
gpt-5.4 to gpt-5.5, the Kimi pairing tagline becomes Kimi K2.6 + GPT-5.5,
and the ultrabrain category now correctly points at GPT-5.5 xhigh.
- Add dedicated Team Mode section after Hephaestus highlighting v4.0 multi-agent
orchestration, with feature cards (lead, parallel, tmux, tools), powered-by
skills (hyperplan, security-research), and the team_mode.enabled opt-in line.
- Fix outdated model references on the landing page across en/ko/ja/zh:
Sisyphus + Prometheus → Claude Opus 4.7 Max, Hephaestus → GPT 5.5 Medium,
Oracle → GPT 5.5 High, Momus → GPT 5.5 xHigh, Metis → Claude Sonnet 4.6,
Librarian + Explore → GPT 5.4 Mini Fast.
- Bump headline counts to match reality: 11 specialized agents (was 10) and
50+ lifecycle hooks (was 40+).
- Replace the 1.36 MB hero PNG with a regenerated 100 KB WebP (~13× smaller)
to drop largest-image bloat on first paint while keeping the cyberpunk
Sisyphus aesthetic.
Sub-agent sessions created via the task tool or call_omo_agent tool were
being created without a model. The resolved model was only passed as a
promptAsync body override, which opencode core ignores, causing fallback
to the client's system default model (often a reasoning model that hangs).
OpenCode's session.create API supports model: { id, providerID, variant }
at creation time. This fix ensures the resolved category/agent model is
passed during session creation across all paths:
- createSyncSession() (delegate task / category-based)
- createOrGetSession() (call_omo_agent / direct agent calls)
- BackgroundManager.startTask() (background tasks)
This guarantees each sub-agent session is created with the correct model
regardless of whether promptAsync honors its model override.
tw-yshuang