feat(anthropic): support Claude Fable 5 adaptive thinking by steipete · Pull Request #91882 · openclaw/openclaw

steipete · 2026-06-10T07:40:13Z

Summary

add Claude Fable 5 to direct Anthropic, Anthropic Vertex, and Amazon Bedrock catalogs/runtimes, including deployment aliases used by Anthropic-compatible transports such as Microsoft Foundry
centralize canonical Claude identity and capability policy so adaptive thinking, effort mapping, replay, catalog, compaction, refusal, and temperature rules do not diverge across providers
enforce Fable's always-adaptive contract: default high; logical off/minimal map to low; route-supported xhigh/max remain available
quarantine partial output until terminal refusal state is known, and reject incomplete terminal streams rather than exposing unsafe output
keep production growth to +1,594/-545 (net +1,049); the larger headline diff is regression coverage across all shared and provider-specific paths

Verification

734 focused regression tests passed across the touched Anthropic, transport, Vertex, Bedrock, catalog, compaction, replay, reasoning, and model-selection surfaces
post-rebase sanity: 167 tests passed across 3 Vitest shards
pnpm build
pnpm plugin-sdk:api:check
fresh autoreview: no actionable findings
AWS Crabbox changed gate: https://crabbox.openclaw.ai/portal/runs/run_61b63b70184e
- core and extension typechecks, lint, tests, import-cycle checks, and architecture guards passed
AWS Crabbox live Anthropic proof: https://crabbox.openclaw.ai/portal/runs/run_902b34e2bb9e
- real openclaw agent --local calls to anthropic/claude-fable-5
- off, minimal, xhigh, and max all returned exact markers with no fallback
GCP Crabbox Vertex probe: https://crabbox.openclaw.ai/portal/runs/run_1a122cd643f7
- source plugin install shape and Fable catalog discovery passed
- live call reached ADC, then stopped because the provisioned VM has no service-account OAuth scopes
AWS Crabbox Bedrock probe: https://crabbox.openclaw.ai/portal/runs/run_773a2c723758
- blocked before model discovery because the runner has no Bedrock identity; both configured local AWS profiles also return invalid-token errors

Real behavior proof

Behavior addressed: Claude Fable 5 works through the canonical Anthropic model policy and direct provider path, with shared behavior for Vertex, Bedrock, and Anthropic-compatible deployment aliases.

Real environment tested: AWS Crabbox Linux with a real Anthropic API credential; GCP Crabbox Linux metadata/ADC path; AWS Crabbox Bedrock discovery path.

Exact steps or command run after this patch: built the synced checkout and ran isolated openclaw agent --local --model anthropic/claude-fable-5 --thinking <level> --json turns for off, minimal, xhigh, and max; separately loaded the Vertex plugin from an installed-plugin-shaped temporary root and attempted a native Bedrock discovery/agent flow.

Evidence after fix: https://crabbox.openclaw.ai/portal/runs/run_902b34e2bb9e

Observed result after fix: all four direct Anthropic calls returned their exact marker; execution metadata reported provider anthropic, model claude-fable-5, terminal stop, and no fallback.

What was not tested: authenticated Vertex, Bedrock, and Microsoft Foundry model calls. No valid Vertex service-account scope, Bedrock credential, or Foundry deployment credential exists in the configured environment; the PR records the exact provider-side blockers above rather than substituting mocks.

Release note

Anthropic: add Claude Fable 5 across direct API, Vertex, Bedrock, and Anthropic-compatible deployment aliases, including adaptive-thinking defaults, replay safety, refusal handling, and canonical model metadata.

clawsweeper · 2026-06-10T07:42:38Z

Codex review: needs maintainer review before merge. Reviewed June 10, 2026, 4:01 AM ET / 08:01 UTC.

Summary
The branch adds Claude Fable 5 catalog metadata and adaptive-thinking/refusal/replay handling across Anthropic direct, Anthropic Vertex, agent reasoning state, docs, and regression tests.

PR surface: Source +491, Tests +1006, Docs +11. Total +1508 across 43 files.

Reproducibility: yes. The linked report provides API-level failure messages for Fable 5, and current main has no Fable-specific adaptive-thinking path, so the failure mode is source-reproducible even though I did not run a live credential repro here.

Review metrics: 2 noteworthy metrics.

Provider proof coverage: 1 live-tested, 1 test-covered. The PR changes both Anthropic direct and Anthropic Vertex behavior; only the direct Anthropic path has submitted live credential proof.
Model catalog surfaces: 2 catalog additions. Adding Fable to both direct Anthropic and Anthropic Vertex catalogs affects model selection and provider metadata visible to users.

Merge readiness
Overall: 🦞 diamond lobster
Proof: 🦞 diamond lobster ✨ media proof bonus
Patch quality: 🦞 diamond lobster
Result: ready for maintainer review.

Overall follows the weaker of proof and patch quality, so missing proof can cap an otherwise strong patch.

Risk before merge

[P1] The protected maintainer label and provider/model-routing scope mean this should remain open for explicit maintainer handling before merge.
[P1] Real behavior proof covers the direct Anthropic provider; Anthropic Vertex support is source- and regression-test-covered but was not exercised with live Vertex credentials.

Maintainer options:

Decide the mitigation before merge
Land this branch after final maintainer review accepts the provider-surface changes and either accepts the test-only Vertex coverage or requests a live Vertex credential proof.
Pause or close
Do not merge this PR until maintainers decide whether the risk is worth taking.

Next step before merge

[P2] Protected maintainer labeling plus provider-surface scope and the Vertex live-proof gap call for explicit maintainer review rather than an automated repair lane.

Security
Cleared: The diff does not touch workflows, lockfiles, dependency sources, package scripts, credentials storage, or secret-handling surfaces; no concrete security or supply-chain concern was found.

Review details

Best possible solution:

Land this branch after final maintainer review accepts the provider-surface changes and either accepts the test-only Vertex coverage or requests a live Vertex credential proof.

Do we have a high-confidence way to reproduce the issue?

Yes. The linked report provides API-level failure messages for Fable 5, and current main has no Fable-specific adaptive-thinking path, so the failure mode is source-reproducible even though I did not run a live credential repro here.

Is this the best way to solve the issue?

Yes, with maintainer review. The PR uses shared provider/model helpers and keeps direct Anthropic, transport, Vertex, replay, and agent reasoning state aligned rather than adding a one-off request patch.

AGENTS.md: found and applied where relevant.

Codex review notes: model gpt-5.5, reasoning high; reviewed against e3fe6715afb6.

Label changes

Label changes:

add P2: This is a normal-priority provider feature/bug-fix path for a newly released Anthropic model with limited blast radius and submitted focused proof.
add rating: 🦞 diamond lobster: Overall readiness is 🦞 diamond lobster; proof is 🦞 diamond lobster and patch quality is 🦞 diamond lobster.
remove rating: 🐚 platinum hermit: Current PR rating is rating: 🦞 diamond lobster, so this older rating label is no longer current.

Label justifications:

P2: This is a normal-priority provider feature/bug-fix path for a newly released Anthropic model with limited blast radius and submitted focused proof.
rating: 🦞 diamond lobster: Overall readiness is 🦞 diamond lobster; proof is 🦞 diamond lobster and patch quality is 🦞 diamond lobster.
status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Sufficient (linked_artifact): The PR body includes a linked AWS Crabbox run with real Anthropic API calls for Fable 5 showing the requested marker for high thinking and logical off, with no fallback; Vertex remains test-covered only.
proof: sufficient: Contributor real behavior proof is sufficient. The PR body includes a linked AWS Crabbox run with real Anthropic API calls for Fable 5 showing the requested marker for high thinking and logical off, with no fallback; Vertex remains test-covered only.

Evidence reviewed

PR surface:

Source +491, Tests +1006, Docs +11. Total +1508 across 43 files.

View PR surface stats

Area	Files	Added	Removed	Net
Source	26	739	248	+491
Tests	15	1032	26	+1006
Docs	2	12	1	+11
Config	0	0	0	0
Generated	0	0	0	0
Other	0	0	0	0
Total	43	1783	275	+1508

What I checked:

Repository policy applied: Read the full root policy plus scoped guides for extensions/, src/agents/, src/agents/embedded-agent-runner/run/, src/plugin-sdk/, src/plugins/, and docs/; provider/model routing, Plugin SDK boundaries, docs, proof, and protected-label guidance affected the review. (AGENTS.md:1, e3fe6715afb6)
Current main lacks Fable support: A current-main search found Anthropic adaptive-thinking support for existing Claude 4.x families but no claude-fable-5 implementation, so the central change is not already implemented on main. (e3fe6715afb6)
Direct Anthropic request shape: At the PR head, Fable requests always enter the adaptive-thinking branch, default effort to high, set thinking: { type: "adaptive" }, and attach output_config.effort. (src/llm/providers/anthropic.ts:1055, 8097d8b09825)
Transport runtime request shape: The Anthropic Messages transport path mirrors the Fable adaptive contract, drops incompatible temperature, normalizes tool choice, and maps logical off/minimal to low effort. (src/agents/anthropic-transport-stream.ts:919, 8097d8b09825)
Vertex and catalog coverage: The PR adds Fable to the Anthropic Vertex catalog, restores required metadata for explicit Vertex rows, and routes Vertex streaming through the same adaptive effort mapping. (extensions/anthropic-vertex/provider-catalog.ts:39, 8097d8b09825)
Agent reasoning state: The PR carries the logical thinking level separately and maps explicit off through the model's thinkingLevelMap.off, which is what makes Fable's logical off become provider-supported low effort instead of disabled thinking. (packages/agent-core/src/reasoning.ts:19, 8097d8b09825)

Likely related people:

steipete: Prior current-main history includes multiple Anthropic docs/model/default and Claude effort commits by Peter Steinberger, including Opus 4.7/xhigh and Anthropic default work; the current PR author also has prior merged history in this area. (role: recent Anthropic feature contributor; confidence: high; commits: a2753e2d9fa6, c73a6d2f689f, 628b454eff6e; files: src/plugin-sdk/provider-model-shared.ts, extensions/anthropic/register.runtime.ts, src/llm/providers/anthropic.ts)
vincentkoc: Recent history shows Vincent Koc centralizing Anthropic replay hooks and provider-model helper behavior that this PR reuses through shared provider/model helpers. (role: provider replay and SDK area contributor; confidence: high; commits: e5c38290a68b, 4c15f1310bd5, e121889a9fed; files: src/plugin-sdk/provider-model-shared.ts, src/plugins/provider-replay-helpers.ts, extensions/anthropic/register.runtime.ts)
jalehman: Josh Lehman authored a recent Anthropic Messages max-token guard commit on the provider runtime path that this PR changes for Fable's 128k output behavior. (role: adjacent max-token behavior contributor; confidence: medium; commits: ef3ac6a58ea3; files: src/llm/providers/anthropic.ts, src/agents/anthropic-transport-stream.ts)

What the crustacean ranks mean

🦀 challenger crab: rare, exceptional readiness with strong proof, clean implementation, and convincing validation.
🦞 diamond lobster: very strong readiness with only minor maintainer review expected.
🐚 platinum hermit: good normal PR, likely mergeable with ordinary maintainer review.
🦐 gold shrimp: useful signal, but proof or patch confidence is still limited.
🦪 silver shellfish: thin signal; proof, validation, or implementation needs work.
🧂 unranked krab: not merge-ready because proof is missing/unusable or there are serious correctness or safety concerns.
🌊 off-meta tidepool: rating does not apply to this item.

Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics.

How this review workflow works

ClawSweeper keeps one durable marker-backed review comment per issue or PR.
Re-runs edit this comment so the latest verdict, findings, and automation markers stay together instead of adding duplicate bot comments.
A fresh review can be triggered by eligible @clawsweeper re-review comments, exact-item GitHub events, scheduled/background review runs, or manual workflow dispatch.
PR/issue authors and users with repository write access can comment @clawsweeper re-review or @clawsweeper re-run on an open PR or issue to request a fresh review only.
Maintainers can also comment @clawsweeper review to request a fresh review only.
Fresh-review commands do not start repair, autofix, rebase, CI repair, or automerge.
Maintainer-only repair and merge flows require explicit commands such as @clawsweeper autofix, @clawsweeper automerge, @clawsweeper fix ci, or @clawsweeper address review.
Maintainers can comment @clawsweeper explain to ask for more context, or @clawsweeper stop to stop active automation.

steipete · 2026-06-10T15:08:27Z

Land-ready verification for 03b58eadefc4dc66afbc0c4317ad383ab8c2c1e9:

GitHub Actions CI run 27285425308: 115 jobs passed, 0 failed.
Crabbox changed gate: run_61b63b70184e — https://crabbox.openclaw.ai/portal/runs/run_61b63b70184e
Direct Anthropic live test: run_902b34e2bb9e — https://crabbox.openclaw.ai/portal/runs/run_902b34e2bb9e
- Actual openclaw agent --local --model anthropic/claude-fable-5 calls passed for off, minimal, xhigh, and max.
- Exact expected markers returned; provider/model/terminal state verified; no fallback.
Vertex: catalog discovery passed and the actual Fable call reached Google ADC in run_1a122cd643f7; the runner service account lacked VM permission scopes.
Bedrock: native discovery/call attempted in run_773a2c723758; available AWS credentials were rejected before model invocation.
Foundry: no Azure endpoint/deployment credential was available in the maintainer environment.
Focused tests: 734 passed before rebase; 167 passed on the clean rebased head.
pnpm build
pnpm plugin-sdk:api:check
extension typecheck/lint
git diff --check
fresh autoreview: no actionable findings

The cloud-route gaps are credential/infrastructure gaps, not mocked substitutions. Direct Anthropic behavior and every supported reasoning level were exercised against the real Fable API before merge.

The re-export at line 6 references resolveClaudeModelIdentity but it was never imported. This was introduced by openclaw#91882 which added the Fable 5 adaptive thinking support and accidentally dropped the import while narrowing the import list. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

…1882) * feat(anthropic): support Claude Fable 5 * test(anthropic): tighten Fable stream fixtures * fix(anthropic): preserve Vertex input types * test(anthropic): use provider-ready Vertex effort * fix(anthropic): support Fable deployment aliases * fix(anthropic): discard incomplete Fable output * feat(anthropic): support Fable on Bedrock * fix(anthropic): preserve Fable reasoning contracts * refactor(anthropic): unify canonical Claude model policy * fix(anthropic): satisfy extension thinking types * test(anthropic): complete canonical alias fixture * fix(bedrock): scope thinking case declarations

…26.6.6) (#1040) This PR contains the following updates: | Package | Update | Change | |---|---|---| | [ghcr.io/openclaw/openclaw](https://openclaw.ai) ([source](https://github.com/openclaw/openclaw)) | patch | `2026.6.5` → `2026.6.6` | --- ### Release Notes <details> <summary>openclaw/openclaw (ghcr.io/openclaw/openclaw)</summary> ### [`v2026.6.6`](https://github.com/openclaw/openclaw/blob/HEAD/CHANGELOG.md#202666) [Compare Source](openclaw/openclaw@v2026.6.5...v2026.6.6) ##### Highlights - Security boundaries are substantially tighter across transcripts, sandbox binds, host environment inheritance, MCP stdio, Codex HTTP access, native search policy, elevated sender checks, deleted-agent ACP bypasses, loopback tools, Discord moderation, and Teams group actions; exec approvals now fail closed on timeout. ([#91529](openclaw/openclaw#91529), [#91618](openclaw/openclaw#91618), [#91615](openclaw/openclaw#91615), [#91619](openclaw/openclaw#91619), [#91741](openclaw/openclaw#91741), [#91745](openclaw/openclaw#91745), [#91746](openclaw/openclaw#91746), [#91748](openclaw/openclaw#91748), [#91749](openclaw/openclaw#91749), [#91750](openclaw/openclaw#91750), [#91751](openclaw/openclaw#91751), [#91752](openclaw/openclaw#91752), [#91763](openclaw/openclaw#91763), [#89938](openclaw/openclaw#89938)) Thanks [@joshavant](https://github.com/joshavant), [@pgondhi987](https://github.com/pgondhi987), [@mmaps](https://github.com/mmaps), [@eleqtrizit](https://github.com/eleqtrizit), [@shakkernerd](https://github.com/shakkernerd), and [@drobison00](https://github.com/drobison00). - Telegram delivery is safer and more coherent: account-scoped topics route to the right agent, streamed text survives tool calls, `/compact` works on generic ingress, callback handling uses concrete APIs, draft chunking is shared, durable dispatch dedupe moved into the SDK, and unauthorized DM text stays out of cache and prompt context. ([#91189](openclaw/openclaw#91189), [#88682](openclaw/openclaw#88682), [#89588](openclaw/openclaw#89588), [#90212](openclaw/openclaw#90212), [#91876](openclaw/openclaw#91876), [#91874](openclaw/openclaw#91874), [#91904](openclaw/openclaw#91904), [#91478](openclaw/openclaw#91478), [#91915](openclaw/openclaw#91915)) Thanks [@codysai001](https://github.com/codysai001), [@alexzhu0](https://github.com/alexzhu0), [@joelnishanth](https://github.com/joelnishanth), [@snowzlm](https://github.com/snowzlm), [@obviyus](https://github.com/obviyus), and [@sallyom](https://github.com/sallyom). - iMessage recovery and delivery now cover always-on inbound restart, durable echo markers, block streaming, idle approval discovery, hardened outbound transport, and actionable inbound startup diagnostics. ([#91335](openclaw/openclaw#91335), [#91449](openclaw/openclaw#91449), [#88969](openclaw/openclaw#88969), [#88530](openclaw/openclaw#88530), [#91783](openclaw/openclaw#91783), [#91785](openclaw/openclaw#91785)) Thanks [@omarshahine](https://github.com/omarshahine), [@jmissig](https://github.com/jmissig), and [@colmbrogan](https://github.com/colmbrogan). - Browser and MCP connectivity gained existing-session CDP support, discovered WebSocket validation, default-profile `cdpUrl` handling, safer browser-output boundaries, Streamable HTTP loopback transport, corrected OAuth/SSE authorization handling, and broader schema compatibility. ([#91422](openclaw/openclaw#91422), [#89851](openclaw/openclaw#89851), [#91736](openclaw/openclaw#91736), [#91747](openclaw/openclaw#91747), [#91451](openclaw/openclaw#91451), [#80143](openclaw/openclaw#80143)) Thanks [@pgondhi987](https://github.com/pgondhi987), [@anagnorisis2peripeteia](https://github.com/anagnorisis2peripeteia), [@lifuyue](https://github.com/lifuyue), [@eleqtrizit](https://github.com/eleqtrizit), [@LiuwqGit](https://github.com/LiuwqGit), and [@HemantSudarshan](https://github.com/HemantSudarshan). - Control UI startup and first-reply latency are lower through cached model metadata, removal of the startup catalog wait, lazy slash-command loading, and first-event tracing with slow-reply diagnostics. ([#91531](openclaw/openclaw#91531), [#91538](openclaw/openclaw#91538), [#91568](openclaw/openclaw#91568), [#91583](openclaw/openclaw#91583), [#91598](openclaw/openclaw#91598)) - Provider support expands with OpenRouter OAuth onboarding and Claude Fable 5 adaptive thinking, while Codex sessions keep correct compaction ownership, local models skip guardian review, dynamic tool progress normalizes cleanly, and Gemma 4 reasoning replay is preserved. ([#91830](openclaw/openclaw#91830), [#91882](openclaw/openclaw#91882), [#91590](openclaw/openclaw#91590), [#88630](openclaw/openclaw#88630), [#88768](openclaw/openclaw#88768), [#91696](openclaw/openclaw#91696)) Thanks [@Patrick-Erichsen](https://github.com/Patrick-Erichsen), [@joshavant](https://github.com/joshavant), [@bdjben](https://github.com/bdjben), and [@Coder-Wangyankun](https://github.com/Coder-Wangyankun). ##### Changes - CLI progress: emit Claude CLI commentary progress events and bridge inter-tool commentary into channel progress without exposing internal protocol scaffolding. ([#89834](openclaw/openclaw#89834), [#90883](openclaw/openclaw#90883)) Thanks [@anagnorisis2peripeteia](https://github.com/anagnorisis2peripeteia). - Observability: allow trusted diagnostics channels to capture tool input/output content, add first-assistant-event traces, and warn on slow initial replies. ([#91256](openclaw/openclaw#91256), [#91568](openclaw/openclaw#91568), [#91583](openclaw/openclaw#91583)) Thanks [@amknight](https://github.com/amknight). - Plugins/ClawHub: dogfood reusable package publishing, let dry runs skip publish approval, allow declared installed trusted hooks, report managed plugin version drift, and warn instead of failing on retired Skill Workshop configuration. ([#91574](openclaw/openclaw#91574), [#91591](openclaw/openclaw#91591), [#90004](openclaw/openclaw#90004), [#90927](openclaw/openclaw#90927), [#90838](openclaw/openclaw#90838)) Thanks [@Patrick-Erichsen](https://github.com/Patrick-Erichsen), [@brokemac79](https://github.com/brokemac79), and [@lonexreb](https://github.com/lonexreb). - Memory/providers: move the local llama.cpp runtime into its provider plugin, batch embeddings across files, persist the agent model catalog cache, and keep QMD JSON search one-shot while filtering stale REM recall previews. ([#91324](openclaw/openclaw#91324), [#89138](openclaw/openclaw#89138), [#90457](openclaw/openclaw#90457), [#91837](openclaw/openclaw#91837), [#91851](openclaw/openclaw#91851)) Thanks [@osolmaz](https://github.com/osolmaz), [@mushuiyu886](https://github.com/mushuiyu886), [@ai-hpc](https://github.com/ai-hpc), and [@TurboTheTurtle](https://github.com/TurboTheTurtle). - Channels/mobile: add the QQBot group mention toggle, improve iPad and iPhone control surfaces, and expose the active connection host in the TUI footer. ([#91423](openclaw/openclaw#91423), [#91557](openclaw/openclaw#91557), [#89909](openclaw/openclaw#89909)) Thanks [@cxyhhhhh](https://github.com/cxyhhhhh), [@Solvely-Colin](https://github.com/Solvely-Colin), and [@baskduf](https://github.com/baskduf). - Performance: prewarm TUI runtime plugins, deduplicate plugin auto-enable fanout, trim dense text-delta snapshots, and reuse prepared startup model metadata. ([#90782](openclaw/openclaw#90782), [#89978](openclaw/openclaw#89978), [#91580](openclaw/openclaw#91580), [#91531](openclaw/openclaw#91531)) Thanks [@RomneyDa](https://github.com/RomneyDa) and [@ai-hpc](https://github.com/ai-hpc). ##### Fixes - Agent/session recovery: drop stale approval follow-ups after session rebind, remove drained reply-queue items by identity, recover stale main and visible replies, preserve Codex context-engine compaction ownership, lower the default compaction timeout to 180 seconds while respecting explicit configuration, and keep provider-failure terminal lifecycle state correct. ([#85679](openclaw/openclaw#85679), [#91450](openclaw/openclaw#91450), [#91566](openclaw/openclaw#91566), [#91840](openclaw/openclaw#91840), [#91590](openclaw/openclaw#91590), [#91361](openclaw/openclaw#91361), [#91895](openclaw/openclaw#91895)) Thanks [@openperf](https://github.com/openperf), [@yetval](https://github.com/yetval), [@joshavant](https://github.com/joshavant), [@wangmiao0668000666](https://github.com/wangmiao0668000666), and [@TurboTheTurtle](https://github.com/TurboTheTurtle). - User-visible content boundaries: suppress Codex/Harmony protocol artifacts, neutralize browser and LanceDB memory media directives, redact transcript images, and preserve native `/compact` replies through source suppression. ([#89151](openclaw/openclaw#89151), [#91422](openclaw/openclaw#91422), [#91425](openclaw/openclaw#91425), [#91529](openclaw/openclaw#91529), [#90212](openclaw/openclaw#90212)) Thanks [@joelnishanth](https://github.com/joelnishanth), [@pgondhi987](https://github.com/pgondhi987), [@joshavant](https://github.com/joshavant), and [@snowzlm](https://github.com/snowzlm). - Channel delivery: keep WhatsApp captured replies attached to the successor controller after restart, retry Feishu rate limits, preserve Mattermost thread replies, canonicalize LINE webhook paths, restore Discord reply hydration and runtime timeout exports, and show OpenAI Realtime WebRTC assistant transcripts. ([#85823](openclaw/openclaw#85823), [#89659](openclaw/openclaw#89659), [#91684](openclaw/openclaw#91684), [#91649](openclaw/openclaw#91649), [#90263](openclaw/openclaw#90263), [#91686](openclaw/openclaw#91686), [#90426](openclaw/openclaw#90426)) Thanks [@itsuzef](https://github.com/itsuzef), [@ladygege](https://github.com/ladygege), [@jacobtomlinson](https://github.com/jacobtomlinson), [@fuller-stack-dev](https://github.com/fuller-stack-dev), and [@shushushv](https://github.com/shushushv). - Cron: cancel active task runs cleanly, preserve terminal timeout/cancel state, and recover no-deliver tool warnings instead of silently losing the outcome. ([#90666](openclaw/openclaw#90666), [#90678](openclaw/openclaw#90678)) Thanks [@ai-hpc](https://github.com/ai-hpc). - Gateway/config/auth: share the approval runtime socket token, replace arrays explicitly in `config.patch`, skip the deleted-agent guard only for valid ACP harness sessions, surface headless LaunchAgent state, verify SQLite auth migration before cleanup, and arm QMD startup maintenance. ([#87105](openclaw/openclaw#87105), [#91551](openclaw/openclaw#91551), [#91219](openclaw/openclaw#91219), [#91614](openclaw/openclaw#91614), [#91740](openclaw/openclaw#91740), [#91978](openclaw/openclaw#91978)) Thanks [@fuller-stack-dev](https://github.com/fuller-stack-dev) and [@scotthuang](https://github.com/scotthuang). - Providers/Codex: clarify quota errors, restore the Codex synthetic usage line, canonicalize Codex protocol assets, require API-key auth for realtime voice, normalize ACP model refs, preserve Gemma 4 `reasoning_content`, and avoid guardian review for local models. ([#91390](openclaw/openclaw#91390), [#91709](openclaw/openclaw#91709), [#91507](openclaw/openclaw#91507), [#91567](openclaw/openclaw#91567), [#88630](openclaw/openclaw#88630), [#91696](openclaw/openclaw#91696)) Thanks [@hxy91819](https://github.com/hxy91819), [@brokemac79](https://github.com/brokemac79), [@RomneyDa](https://github.com/RomneyDa), [@joshavant](https://github.com/joshavant), and [@Coder-Wangyankun](https://github.com/Coder-Wangyankun). - Updates/builds: recover package Gateway restarts after refresh failure, expose plugin convergence repair, fall back to Corepack in PATH-less pnpm environments, seed the correct Docker store packages, and keep ClawHub dry-run and publish paths reusable. ([#91581](openclaw/openclaw#91581), [#91599](openclaw/openclaw#91599), [#91547](openclaw/openclaw#91547), [#91591](openclaw/openclaw#91591)) Thanks [@fuller-stack-dev](https://github.com/fuller-stack-dev), [@sallyom](https://github.com/sallyom), and [@Patrick-Erichsen](https://github.com/Patrick-Erichsen). - UI: require explicit user intent before opening chat sessions and drain restored chat queues after session switches. ([#91480](openclaw/openclaw#91480)) Thanks [@TurboTheTurtle](https://github.com/TurboTheTurtle). - Android: avoid the `dataSync` foreground-service type for persistent nodes. ([#80082](openclaw/openclaw#80082)) Thanks [@davelutztx](https://github.com/davelutztx). - Native hooks: bound relay lifetimes so abandoned native hook connections cannot linger indefinitely. ([#91550](openclaw/openclaw#91550)) Thanks [@joshavant](https://github.com/joshavant). </details> --- ### Configuration 📅 **Schedule**: Branch creation - At any time (no schedule defined), Automerge - At any time (no schedule defined). 🚦 **Automerge**: Disabled by config. Please merge this manually once you are satisfied. ♻ **Rebasing**: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox. 🔕 **Ignore**: Close this PR and you won't be reminded about these updates again. --- - [ ] If you want to rebase/retry this PR, check this box --- This PR has been generated by [Renovate Bot](https://github.com/renovatebot/renovate).  Reviewed-on: https://git.erwanleboucher.dev/eleboucher/homelab/pulls/1040

openclaw-barnacle Bot added docs Improvements or additions to documentation agents Agent runtime and tooling extensions: anthropic extensions: anthropic-vertex size: XL maintainer Maintainer-authored PR labels Jun 10, 2026

steipete self-assigned this Jun 10, 2026

steipete added 12 commits June 10, 2026 15:07

feat(anthropic): support Claude Fable 5

63b81ac

test(anthropic): tighten Fable stream fixtures

84b736a

fix(anthropic): preserve Vertex input types

73bc4e7

test(anthropic): use provider-ready Vertex effort

1df08ed

fix(anthropic): support Fable deployment aliases

126dfbf

fix(anthropic): discard incomplete Fable output

5a39002

feat(anthropic): support Fable on Bedrock

38e40d3

fix(anthropic): preserve Fable reasoning contracts

cc42490

refactor(anthropic): unify canonical Claude model policy

9b695ca

fix(anthropic): satisfy extension thinking types

5e37ea3

test(anthropic): complete canonical alias fixture

a8aefd7

fix(bedrock): scope thinking case declarations

03b58ea

steipete force-pushed the fix/anthropic-fable-5 branch from 8097d8b to 03b58ea Compare June 10, 2026 15:00

openclaw-barnacle Bot added extensions: amazon-bedrock extensions: github-copilot labels Jun 10, 2026

steipete merged commit 0e7b5c3 into main Jun 10, 2026
179 of 181 checks passed

steipete deleted the fix/anthropic-fable-5 branch June 10, 2026 15:08

github-actions Bot mentioned this pull request Jun 10, 2026

📡 Upstream Digest — 2026-06-10 16:54 UTC curtismercier/openclaw-mods#1053

Open

fsdwen mentioned this pull request Jun 11, 2026

fix(anthropic): resolve thinking profile for custom Anthropic-compatible providers #92071

Closed

clawsweeper Bot mentioned this pull request Jun 11, 2026

Feature: Support custom anthropic_beta flags in Bedrock model config #39734

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

feat(anthropic): support Claude Fable 5 adaptive thinking#91882

feat(anthropic): support Claude Fable 5 adaptive thinking#91882
steipete merged 12 commits into
mainfrom
fix/anthropic-fable-5

steipete commented Jun 10, 2026 •

edited

Loading

Uh oh!

clawsweeper Bot commented Jun 10, 2026 •

edited

Loading

Uh oh!

steipete commented Jun 10, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Uh oh!

Conversation

steipete commented Jun 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Verification

Real behavior proof

Release note

Uh oh!

clawsweeper Bot commented Jun 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

steipete commented Jun 10, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

steipete commented Jun 10, 2026 •

edited

Loading

clawsweeper Bot commented Jun 10, 2026 •

edited

Loading