OMO 5.0.0-beta.8

This release is about doing more work at once and trusting what runs in the background. OMO can now run dozens of agents in parallel and recover if something crashes mid-run. You can sign in with your Cursor subscription, and Grok 4.6 is supported out of the box. The background memory system, which learns from your conversations, got a major reliability overhaul after an incident where it filled a disk with retries. Rounding it out: your settings now survive every update, planning asks fewer questions, and Windows got smoother.

Highlights

Run dozens of agents at once

Give OMO a big task and it can now break it into parts, run them as parallel waves of agents, collect the results, and move on to the next wave. Before, agents mostly ran one at a time. If a crash happens partway through, OMO picks up where it left off instead of losing the work. Nothing to configure, big tasks just fan out on their own.

Sign in with Cursor, and Grok 4.6 support

You can now connect your Cursor Pro, Ultra, or Teams subscription. Run /login cursor, approve in your browser, and you're signed in. For now this is authentication only. Cursor models don't show up in the model picker yet, but the groundwork is in place.

Grok 4.6 is now a first-class model. It gets a system prompt tuned for how Grok actually works instead of a generic fallback, and when OMO picks a model for quick, lightweight tasks, Grok 4.6 is the new default.

Memory that doesn't eat your disk

OMO quietly extracts facts from your conversations so it remembers things about you and your projects. On August 14 that system hit an infinite retry loop and consumed 388GB of disk. This release fixes all five root causes:

• Failed extractions back off and stop retrying after a few attempts instead of looping forever.
• Disk usage is bounded. Old extraction runs are cleaned up automatically and payloads are capped at 128KiB.
• You can check what the memory system is doing with /facts status and retry a failed extraction with /facts retry.
• Background memory consolidation no longer silently fails on many model providers. If its preferred model isn't available, it falls back to whatever is.
• Model names with slashes in them (like openrouter/deepseek-ai/deepseek-v4-pro) are no longer invisible to the memory system.

There's new smarts here too. Memory consolidation now decides between reusing your active session (faster, cache-warm) and starting fresh (more reliable) based on actual token cost. And the planning review loop can no longer run indefinitely; it's capped at 5 rounds.

Your settings survive every update

Updating OMO used to sometimes wipe your favorite models, fallback chains, or auth tokens, because different parts of the system looked for settings in different places. Now there's one home for your config, ~/.omo/agent, and everything (launchers, child agents, doctor, setup, the installer) uses it. A one-time migration moves your existing settings there automatically.

Smarter planning when you don't spell everything out

When you ask for something without specifying budget, tech stack, scale, or compliance needs, OMO used to either assume nothing or interrogate you. Now it derives sensible defaults, records what it assumed so you can review it, and asks only the one question that genuinely needs your input.

See how much time parallel work saves

OMO now measures the wall-clock time saved by running tool calls in parallel and reports it. The old standalone /omo-telemetry command is removed; the measurement is built in now.

Smoother on Windows

Spawning a background agent on Windows no longer flashes a console window, thanks to community contributor @sanguneo. A batch of timeout and race condition fixes landed across Windows and macOS as well.

Agents doing UI work now actually study design

OMO ships a library of design references for frontend work, but agents mostly skimmed past it: across thousands of real sessions, only about 2 in 10 ever opened it, and almost none followed through. The router now makes the reference stops binding instead of advisory, adds a dedicated layout-mechanics reference, and requires the agent to name which references it actually loaded before writing UI code. Expect more design-grounded frontends, fewer generic ones.

Fresher web automation under the hood

The bundled browser automation stack (the stealth browser and its controller) is refreshed to the current releases, with a newer Playwright baseline. Web research, login-gated scraping, and screenshot workflows track modern browser behavior instead of last quarter's.

Team Mode gets boundary repairs

Team Mode, the experimental parallel-agent coordination feature, had runtime boundary defects where members could step outside their assigned scope. Those boundaries are repaired, and read-only shell output is now capped so a chatty command can no longer blow up an agent's context window.

Edition impact

OMO Ultimate for OpenCode

The frontend design routing and browser automation refresh land here first. Grok 4.6 becomes the default for quick tasks, planning derives constraints instead of asking, and the memory reliability fixes apply through the shared memory core. Your settings now live in the canonical config directory and survive updates.

OMO Light for Codex

The planning constraint improvements reach the Codex skills, CodeGraph is refreshed, and the Git Bash installer is hardened. No new Codex hooks, tools, or CLI commands this release.

OMO Native

Everything above, plus the full Senpi engine upgrade below: Cursor sign-in, Grok 4.6, JSONC settings, mass parallel orchestration, parallelism telemetry, the memory disk fix with the new /facts commands, and the Windows console flash fix. The /omo-telemetry command is removed.

Senpi engine upgrade to 2026.8.16

Senpi is the engine that powers OMO Native (the omo-ai package). This release upgrades it from 2026.8.14 to 2026.8.16.

New things you can do:

• Sign in with Cursor. /login cursor opens your browser, you approve with your Cursor account, and you're connected. Authentication only for now.
• Grok 4.6 tuned prompt. Grok 4.6 gets a system prompt designed for how it thinks instead of a generic one.
• Custom system prompts work again. The --system-prompt and --append-system-prompt flags are restored after being broken since July 19, and they compose correctly with model-specific presets.
• JSONC settings. Your settings file can have comments and trailing commas. If both settings.jsonc and settings.json exist, JSONC wins.
• GLM 5.3 support. GLM 5.3 models get a tuned system prompt too.
• Per-model reasoning memory. OMO remembers each model's thinking level and fast-tier setting across sessions. New /reasoning and /efforts commands let you check and change them, and /fast now persists.
• Dollar skill invocation. Type $skill-name in the composer to run a skill directly.
• Eval execution events are now published for desktop clients.

Things that are fixed:

• Background tasks no longer crash in restricted environments where they can't create a lock file. They degrade gracefully instead.
• Aborted or oversized conversation compactions no longer freeze your session, and when a compaction doesn't apply, you're told why.
• Claude authentication is more reliable: tokens stay scoped to the right request, are isolated from subprocesses, and the auth probe window no longer flashes on macOS.

Migration note for users coming from 4.x

The omo to omo-agent-toolkit rename from the 5.0 beta line remains in effect:

omo ...  ->  omo-agent-toolkit ...

Rerunning the installer removes only installer-managed legacy wrappers and preserves any omo files you created yourself.

Release surfaces

• oh-my-opencode@5.0.0-beta.8
• oh-my-openagent@5.0.0-beta.8
• lazycodex-ai@5.0.0-beta.8
• omo-ai@5.0.0-0.beta.8 (beta channel only: npm i -g omo-ai@beta)
• Codex marketplace plugin omo@sisyphuslabs
• platform launcher packages stamped 5.0.0-beta.8

Thanks to community contributor @sanguneo for the Windows console flash fix.

Full diff: v5.0.0-beta.7...v5.0.0-beta.8

• ea02aa0 Merge pull request #6955 from code-yeongyu/release/v5.0.0-beta.8-source-state
• 43b5151 Merge pull request #6956 from code-yeongyu/fix/release-pr-read-retries
• 600089d fix(ci): retry every release-state PR read on HTTP 5xx
• 988c753 Merge pull request #6949 from code-yeongyu/fix/release-wait-loop-resilience
• 1d25873 fix(ci): retry gh pr view on transient HTTP 5xx in release-state wait loop
• 581881f Merge pull request #6925 from code-yeongyu/fix/ci-real-sharding
• 917e29d Merge pull request #6943 from code-yeongyu/fix/sdk-message-lookup-non-array
• 001d7ec Merge pull request #6947 from code-yeongyu/fix/quick-luna-fast-provider
• 6c1ff16 Merge pull request #6946 from code-yeongyu/fix/release-gates
• 08a10e6 Merge pull request #6941 from code-yeongyu/fix/rpc-parity-local-cli
• 0543401 Merge pull request #6939 from code-yeongyu/fix/lazycodex-parallel-fixture
• 010ff3f fix(model-core): use openai-codex provider for quick luna-fast rung
• 3f260e8 fix(memory-core): retry git init under lock contention
• bbebf5e test(opencode): cover resolved message lookup failures
• 557ff51 fix(opencode): preserve array SDK response contracts
• dbf6005 Merge pull request #6933 from code-yeongyu/feat/publish-post-verify
• e3b6860 Merge pull request #6945 from code-yeongyu/fix/windows-dag-happy-timeout
• 5b02fd4 docs(evidence): record release-gate repair verification
• 469b058 test(omo-codex): isolate LazyCodex installer source fixtures
• 0b025ec Merge pull request #6942 from code-yeongyu/test/windows-dag-residency-budget
• 397d534 Merge pull request #6936 from code-yeongyu/docs/release-resume-path
• 89a37db Merge pull request #6940 from code-yeongyu/fix/publish-rpc-parity-launcher
• b57dcd8 Merge pull request #6928 from code-yeongyu/perf/slow-test-seams
• 05ef183 fix(publish): stop post-publish probes from stranding a release
• d9cdf02 Merge pull request #6935 from code-yeongyu/feat/publish-gate-reuse
• 4c4553f docs(release): document failed publish recovery
• f057c7e ci(publish): reuse successful release gates
• c2895ee test(release): pin publish resume guards
• 105258b Merge pull request #6932 from code-yeongyu/fix/frontend-reference-enforcement
• 386975a Merge pull request #6920 from code-yeongyu/fix-beta8-command-audit
• 2509bfc Merge pull request #6931 from code-yeongyu/fix/test-parallel-worker-isolation
• 2ff7f7d fix(frontend): route layout mechanics beside the pattern catalog
• aa57607 fix(frontend): make routed references binding instead of advisory
• b280c3f feat(frontend): add StyleGallery spatial-structure reference
• 2925acc fix(audit): make the command-string allowlist line-number-free
• 97e0ec3 test(lsp-core): stabilize cold diagnostics delivery
• 04ccd59 Merge pull request #6924 from code-yeongyu/chore/dep-pins-latest
• 1ef6b28 docs(evidence): attribute the Windows shard 2/2 attempt-1 failure
• 7c239c0 docs(evidence): record post-review verification for the pin refresh
• 32f380f fix(script): correct stale agent-command allowlist line pins
• 1ade4ef chore(deps): advance tsgo to the current native-preview release
• 8bd0dee chore(shared-skills): refresh ultimate-browsing Tier-2 runtime pins
• 3dd8826 docs: refresh knowledge base snapshot and add component guides
• 1f87b56 test(memory): stabilize time-sensitive CI cases
• 7921e9c docs(evidence): record executable path hardening
• 82f3aeb docs(evidence): record final root verification
• ff46946 test(team-mode): avoid global fake timers
• b89272b docs(evidence): record security blocker repairs
• 01eb5fb docs(evidence): record final Windows proof
• c69a294 test(memory-core): budget Windows cache integration
• 174b317 docs(evidence): record Windows CI root causes
• ce3ff30 test(memory): inject terminal lock record fixture
• 0947b50 test(init-deep): build drift history with fast-import
• 5d097ae fix(memory-core): serialize worktree administration
• ee63da5 docs(evidence): record brutal review repair
• 8ce91d0 test(omo-codex): bound Git Bash installer fixture
• 26cc996 Merge pull request #6913 from code-yeongyu/feat/unspecified-low-grok-4.6
• 9d0536a fix: keep grok-4.6 snapshot-backed on unspecified-low
• 8825a9c docs: document unspecified-low grok-4.6 first lane
• c86e34a feat(omo-opencode): default unspecified-low to xai/grok-4.6 xhigh
• 828c252 feat(model-core): put grok-4.6 xhigh first on unspecified-low
• 8eca0d0 Merge pull request #6906 from code-yeongyu/feat/memory-facts-payload-cap
• aeddcb5 Merge pull request #6910 from code-yeongyu/cursor/grok-sisyphus-prompt-4c49
• 19ea2c3 feat(memory-core): lossless byte-capped facts batch selection
• 03b7df0 Merge pull request #6900 from code-yeongyu/feat/memory-facts-retention
• 0661a22 Merge pull request #6899 from code-yeongyu/feat/memory-facts-backoff
• bd0dc3e fix(memory-core): keep parked facts failures parked and reject non-ISO instants
• d17b2fd feat(memory-core): durable facts failure-streak store
• feeaede Merge pull request #6901 from code-yeongyu/fix/memory-journal-lock-contention
• daa2310 fix(memory): make transcript journal lock crash-proof against contention
• 40c4e04 Merge remote-tracking branch 'origin/dev' into cursor/grok-sisyphus-prompt-4c49
• dcd43ed Merge pull request #6897 from code-yeongyu/feat/telemetry-parallel-latency
• a725cbc docs(plan): record completed plan checkboxes for parallelism telemetry
• 2ae457c docs(evidence): record verification verdicts and final QA artifacts for parallelism telemetry
• fe9bd2e docs(evidence): record post-commit bundle verification transcript
• c1b64ca Merge pull request #6909 from code-yeongyu/fix/reflection-completion-stale-fixture
• 4586c62 docs(evidence): record Grok Sisyphus prompt QA evidence
• a88f1e2 feat(sisyphus): route Grok 4.5/4.6 to a native model-specific prompt
• 7ab5376 feat(model-core): add Grok 4.5/4.6 family detectors
• 0512e0c test(memory): use a module-level recent timestamp instead of a hardcoded date in completion fixtures
• 3cb1d63 Merge pull request #6904 from code-yeongyu/feat/ulw-plan-constraint-sweep
• 004fe9d Merge pull request #6903 from code-yeongyu/feat/ulw-loop-assumed-constraints
• 381738a feat(ulw-plan): sweep extrinsic constraints (budget/stack/scale/audience) across editions
• 1301238 feat(ulw-loop): derive assumed constraints when the brief is silent
• bd9987e Merge pull request #6892 from code-yeongyu/fix/memory-facts-sandbox
• c8614e6 Merge pull request #6896 from code-yeongyu/fix/windows-soul-watermark-timeout
• 9e86f63 Merge pull request #6865 from code-yeongyu/feat/mass-ulw-dag-orchestration
• 0360dae test(utils): extend codegraph upgrade test timeout
• e00eaef test(memory): stabilize Windows RPC cleanup
• c58d45e Merge remote-tracking branch 'origin/dev' into feat/mass-ulw-dag-orchestration
• 8ff6182 test(init-deep): avoid Git drift fixtures
• ec7d53e Merge remote-tracking branch 'origin/dev' into fix/windows-soul-watermark-timeout
• 4751bb1 Merge pull request #6893 from code-yeongyu/fix/remove-omo-telemetry-command
• 5c3110e docs(evidence): record criterion-named Team Mode QA artifacts
• 429556b docs(evidence): record Team Mode repair QA evidence
• 589c336 test(memory): harden soul watermark timeout
• 7a15e51 Merge remote-tracking branch 'origin/dev' into feat/mass-ulw-dag-orchestration
• 27fc628 docs(evidence): commit full-suite transcript for telemetry command removal
• 95fe3be Merge pull request #6890 from code-yeongyu/fix/ci-optimization
• 79e5e3c test(memory): harden memory tool and integration tests for windows CI
• 84e3381 Merge pull request #6891 from code-yeongyu/fix/work-with-pr-monitor
• 775cb95 Merge pull request #6889 from code-yeongyu/fix/6881-memory-prompt-cache-stability
• c039370 Merge remote-tracking branch 'origin/dev' into fix/6881-memory-prompt-cache-stability
• 25cb7c7 Merge pull request #6886 from code-yeongyu/fix/plan-gate-user-request-channels
• 76a6d5f fix(ci): build full codex plugin for platform smoke tests
• 63d873b docs(work-with-pr): replace polling loops with monitor subscribe
• 578c25b Merge remote-tracking branch 'origin/dev' into feat/mass-ulw-dag-orchestration
• 15ad6e6 Merge pull request #6885 from code-yeongyu/fix/windows-reflection-fork-test-timeout
• bdc712f Merge remote-tracking branch 'origin/dev' into fix/plan-gate-user-request-channels
• f5959d2 Merge remote-tracking branch 'origin/dev' into fix/6881-memory-prompt-cache-stability
• 6f04a34 Merge pull request #6884 from code-yeongyu/fix/memory-model-catalog-slash-ids
• b448cfe test(memory): harden command fixture timeouts
• 58cfd74 Merge remote-tracking branch 'origin/dev' into fix/6881-memory-prompt-cache-stability
• 0664539 Merge remote-tracking branch 'origin/dev' into fix/memory-model-catalog-slash-ids
• 1b593cf Merge pull request #6888 from code-yeongyu/fix/6883-quick-chain-qwen-flash
• 4296367 Merge remote-tracking branch 'origin/dev' into fix/plan-gate-user-request-channels
• aa7e475 Merge remote-tracking branch 'origin/dev' into fix/windows-reflection-fork-test-timeout
• 3e324d9 docs(evidence): record memory prompt cache QA
• 472d1e2 fix(memory): stabilize compiled prompt projection
• 98f0839 test(memory-core): stop racing git worktree add in the concurrency test
• 51b5028 Merge remote-tracking branch 'origin/dev' into fix/6883-quick-chain-qwen-flash
• eca1d87 Merge remote-tracking branch 'origin/dev' into fix/memory-model-catalog-slash-ids
• f2956f9 Merge pull request #6887 from code-yeongyu/fix/6880-rpc-child-error-classification
• 6e583b3 test(memory): pin the shared launch preflight for both memory surfaces
• 56a0f52 docs: align quick fallback providers
• 75b82be fix(model-core): correct OpenCode Go quick fallback
• 32cc219 test(memory): harden reflection fork timeouts
• 3536c86 Merge remote-tracking branch 'origin/dev' into feat/mass-ulw-dag-orchestration
• 2d55567 fix(memory): name the model when the child reports it cannot see one
• 6cdb81d fix(memory): keep slash-containing model ids visible to preflight
• 8726d98 test(memory): harden Windows process tests
• 0684852 fix(dag): defer in-process teardown past cancellation
• 598bd77 fix(dag): wire shipped handles and recovery cancellation
• 24db42b fix(dag): compose cancellation runtime surfaces
• 6974ab6 fix(dag): avoid viewer protocol fsync cost
• 1d2d814 fix(dag): close final conformance remainders
• c3a0221 Merge remote-tracking branch 'origin/dev' into feat/mass-ulw-dag-orchestration
• 108bffa fix(script): pin dag eval sdk in omo-ai payload guard
• 038ed0c Merge pull request #6864 from code-yeongyu/feat/memory-fork-cost-routing
• 3122f56 fix(memory): judge fork cache reuse from session usage, not assumption
• 6ab5399 feat(memory): route reflection between fork and quick by cost
• be73200 feat(memory): model fork-vs-quick launch cost per turn
• b2273e9 fix(memory): honor the child sentinel as a hard disable
• 28f184c Merge pull request #6859 from code-yeongyu/fix/memory-reflection-runtime-model
• 81a21ec Merge remote-tracking branch 'origin/dev' into fix/memory-reflection-runtime-model
• 3bbdf60 Merge pull request #6860 from code-yeongyu/fix/omo-agent-dir-unification
• 6de0f2d test(memory): cover the session seam and registry scan
• a047dcb test(memory): pin the resolution ladder order
• 6bd290b fix(omo-native): type the adoption result and categorize the new command strings
• f612eb6 docs(evidence): record the canonical agent-directory QA
• 5254268 docs: state the canonical omo agent directory and its invariant
• ee71abf fix(memory): point category failures at the real fix
• 46a3ebe fix(memory): resolve reflection models from the live runtime
• 0d1505b feat(omo-native): add the shared agent-directory resolver
• 259fd76 Merge pull request #6856 from code-yeongyu/fix/ulw-plan-bounded-review-convergence
• 13a034b fix(ulw-plan): bound high-accuracy review convergence
• 8e71c1b Merge pull request #6851 from code-yeongyu/fix/windows-facts-reservation-timeout
• 29ec975 test(memory): budget reservation finalization on Windows
• ac65040 Merge pull request #6839 from code-yeongyu/fix/macos-tmux-manager-timer-isolation
• 9cfe40b Merge origin/dev into fix/macos-tmux-manager-timer-isolation
• 6d206e1 Merge pull request #6838 from code-yeongyu/fix/windows-agent-toolkit-timeout
• b8ae9f7 Merge pull request #6837 from code-yeongyu/fix/windows-port-auto-selection-race
• 4d951e3 test(omo-opencode): isolate tmux polling timers
• 9ab4cd9 test(utils): make port auto-selection deterministic
• 30d7ed5 test(memory): bound cache variant coverage
• 6d3c24f test(omo-native): verify provider registry map
• c26b8fe Merge pull request #6825 from code-yeongyu/fix/memory-core-git-lock-contention
• 1bc6dd6 Merge remote-tracking branch 'origin/dev' into fix/memory-core-git-lock-contention
• ca30fac Merge pull request #6824 from code-yeongyu/fix/memory-journal-stale-lock
• ee11fba test(memory): pin that an unclearable lock still fails loudly
• 564d6da docs(memory): state which proof covers which criterion
• b4b6aa2 test(memory): assert every concurrent worktree commit lands
• 82efbe7 docs(memory): record the git lock contention evidence
• 164dca8 fix(memory): retry the remaining lock-taking git mutations
• dddcbd2 fix(memory): apply the git lock retry to mutating repo operations
• 3d03d1c fix(memory): retry git operations that lose a lock race
• a93f438 fix(memory): use the real commit author shape in the contention probe
• ea82b14 test(memory): probe concurrent git mutations for Windows lock contention
• 6281e9c fix(memory): reclaim stale transcript journal locks from dead owners
• 00bd5b2 Merge pull request #6821 from code-yeongyu/refactor/memory-module-split-runner
• 9c077f2 fix(memory): make runner deadline assertion deterministic
• cb96c44 Merge remote-tracking branch 'origin/dev' into refactor/memory-module-split-runner
• 757f506 Merge pull request #6817 from code-yeongyu/feat/memory-tui-entry-polish
• 378d560 Merge remote-tracking branch 'origin/dev' into feat/memory-tui-entry-polish
• f0d8e1d refactor(memory): split runner and sandbox modules
• 7a00a41 Merge pull request #6812 from code-yeongyu/refactor/memory-health-alert-split
• dc878de Merge remote-tracking branch 'origin/dev' into feat/memory-tui-entry-polish
• f59bec6 Merge remote-tracking branch 'origin/dev' into refactor/memory-health-alert-split
• 55aaa31 Merge pull request #6814 from code-yeongyu/refactor/memory-module-split-bridge
• 703d11d Merge remote-tracking branch 'origin/dev' into feat/memory-tui-entry-polish
• 9008e11 Merge remote-tracking branch 'origin/dev' into feat/memory-tui-entry-polish
• 0b39fa3 Merge pull request #6816 from code-yeongyu/fix/omo-ai-bun-global-install
• b20c141 Merge pull request #6820 from code-yeongyu/fix/format-beta7-changelog
• 2eea2f9 fix: format the beta.7 changelog entry
• de6e55a Merge pull request #6819 from code-yeongyu/docs/record-beta7-release
• 74b7004 docs: record the OmO beta.7 release
• 4c59f88 Merge remote-tracking branch 'origin/dev' into refactor/memory-module-split-bridge
• 5295ad3 Merge remote-tracking branch 'origin/dev' into refactor/memory-health-alert-split
• 4e3c361 Merge pull request #6811 from code-yeongyu/fix/memory-sandbox-escape-visibility
• c2b2375 fix(omo-native): honor Windows runtime home for migration
• e71d0fa Merge remote-tracking branch 'origin/dev' into refactor/memory-module-split-bridge
• e0961dc Merge remote-tracking branch 'origin/dev' into refactor/memory-health-alert-split
• dd4986b Merge remote-tracking branch 'origin/dev' into fix/memory-sandbox-escape-visibility
• 530c2c3 Merge pull request #6815 from code-yeongyu/refactor/memory-module-split-wiring
• b8996a6 Merge remote-tracking branch 'origin/dev' into feat/memory-tui-entry-polish
• ceb9d4a docs(qa): record before/after evidence for the memory entry restyle
• 0398261 test(memory): pin the rendered shape and semantic colour of memory entries
• 2619e8c fix(omo-native): repair legacy Bun global manifests
• 073cfde Merge remote-tracking branch 'origin/dev' into fix/memory-sandbox-escape-visibility
• 25cd1d1 Merge remote-tracking branch 'origin/dev' into refactor/memory-module-split-wiring
• e65d761 Merge remote-tracking branch 'origin/dev' into refactor/memory-module-split-bridge
• 628c8e2 Merge remote-tracking branch 'origin/dev' into refactor/memory-health-alert-split
• d1b1be6 Merge pull request #6810 from code-yeongyu/refactor/memory-shared-launch-preflight
• 25ee0b9 fix(memory): label facts sandbox warnings
• 61f6247 refactor(memory): split session wiring responsibilities
• 91cca1d refactor(memory): split rpc and completion modules
• 38baf46 docs(evidence): record health alert split and spinner assertion proof
• f6db62e docs(qa): record sandbox visibility evidence
• c93efdc refactor(memory): share launch preflight
• c83c62a fix(memory): surface sandbox degradation
• ca64bff test(memory): pin literal spinner glyphs instead of deriving from the frame table
• dad9472 refactor(memory): move reflection health alert out of read-only health module

Thank you to 1 community contributor:

• @cursoragent:
  • feat(model-core): add Grok 4.5/4.6 family detectors
  • feat(sisyphus): route Grok 4.5/4.6 to a native model-specific prompt
  • docs(evidence): record Grok Sisyphus prompt QA evidence
  • Merge remote-tracking branch 'origin/dev' into cursor/grok-sisyphus-prompt-4c49