content large_stringlengths 3 20.5k | url large_stringlengths 54 193 | branch large_stringclasses 4
values | source large_stringclasses 42
values | embeddings listlengths 384 384 | score float64 -0.21 0.65 |
|---|---|---|---|---|---|
## Offline regression (CI-safe) These are “real pipeline” regressions without real providers: - Gateway tool calling (mock OpenAI, real gateway + agent loop): `src/gateway/gateway.test.ts` (case: "runs a mock OpenAI tool call end-to-end via gateway agent loop") - Gateway wizard (WS `wizard.start`/`wizard.next`, writes ... | https://github.com/openclaw/openclaw/blob/main//docs/help/testing.md | main | opebclaw | [
-0.12113571912050247,
-0.030018150806427002,
-0.08611880242824554,
0.11664905399084091,
0.042814843356609344,
-0.028499815613031387,
-0.009167336858808994,
0.0019028724636882544,
-0.05198553204536438,
-0.0006895104888826609,
-0.022659696638584137,
0.0043397205881774426,
0.03588130325078964,
... | 0.079489 |
registry metadata (`listSecretTargetRegistryEntries()`), then asserts traversal-segment exec ids are rejected. - If you add a new `includeInPlan` SecretRef target family in `src/secrets/target-registry-data.ts`, update `classifyTargetClass` in that test. The test intentionally fails on unclassified target ids so new cl... | https://github.com/openclaw/openclaw/blob/main//docs/help/testing.md | main | opebclaw | [
-0.11869774758815765,
0.007008706219494343,
-0.022374605759978294,
0.020586196333169937,
0.04205387085676193,
-0.039577316492795944,
0.02467171661555767,
-0.030393753200769424,
-0.07764159142971039,
0.005917852744460106,
0.0449250303208828,
-0.05201547220349312,
-0.054235491901636124,
0.01... | 0.011527 |
# GPT-5.4 / Codex Agentic Parity in OpenClaw OpenClaw already worked well with tool-using frontier models, but GPT-5.4 and Codex-style models were still underperforming in a few practical ways: - they could stop after planning instead of doing the work - they could use strict OpenAI/Codex tool schemas incorrectly - the... | https://github.com/openclaw/openclaw/blob/main//docs/help/gpt54-codex-agentic-parity.md | main | opebclaw | [
-0.0847209095954895,
-0.06955893337726593,
-0.03621712699532509,
0.023779109120368958,
0.028158433735370636,
-0.09111277014017105,
-0.10677992552518845,
0.011887399479746819,
-0.044499751180410385,
0.0312090702354908,
-0.040671464055776596,
0.038663845509290695,
-0.03579067811369896,
0.002... | -0.004942 |
either acted, or OpenClaw surfaced the exact reason it could not” ## Before vs after for GPT-5.4 users | Before this program | After PR A-D | | ---------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------... | https://github.com/openclaw/openclaw/blob/main//docs/help/gpt54-codex-agentic-parity.md | main | opebclaw | [
-0.09202133119106293,
0.030548976734280586,
0.002613029908388853,
-0.006678025703877211,
0.03715971112251282,
-0.06401178240776062,
-0.00006572582060471177,
0.05132767930626869,
0.01166269090026617,
0.09264929592609406,
0.06243659183382988,
0.04901621863245964,
-0.039118945598602295,
-0.00... | -0.019191 |
the attachment, connects it to tools, and continues the task | vague narration, attachment ignored, or no concrete next action | | `compaction-retry-mutating-tool` | Mutating work under compaction pressure | Performs a real write and keeps replay-unsafety explicit after the side effect | mutating write happens but repl... | https://github.com/openclaw/openclaw/blob/main//docs/help/gpt54-codex-agentic-parity.md | main | opebclaw | [
-0.05632134526968002,
-0.11590096354484558,
0.013306654058396816,
0.016304239630699158,
-0.018418433144688606,
-0.02833530865609646,
-0.06861727684736252,
0.029915092512965202,
0.017553456127643585,
-0.01397029310464859,
0.047408875077962875,
0.04343130439519882,
-0.07629022002220154,
-0.0... | 0.120361 |
looser behavior - you are not using GPT-5-family models - you are testing prompts rather than runtime enforcement | https://github.com/openclaw/openclaw/blob/main//docs/help/gpt54-codex-agentic-parity.md | main | opebclaw | [
-0.04735584184527397,
-0.03244614228606224,
0.03148360177874565,
0.012451655231416225,
0.09578834474086761,
-0.0991913452744484,
-0.013685538433492184,
-0.0013813716359436512,
-0.049165260046720505,
0.017424317076802254,
0.015775345265865326,
-0.016668085008859634,
-0.022493278607726097,
0... | -0.061468 |
# Help If you want a quick “get unstuck” flow, start here: - \*\*Troubleshooting:\*\* [Start here](/help/troubleshooting) - \*\*Install sanity (Node/npm/PATH):\*\* [Install](/install/node#troubleshooting) - \*\*Gateway issues:\*\* [Gateway troubleshooting](/gateway/troubleshooting) - \*\*Logs:\*\* [Logging](/logging) a... | https://github.com/openclaw/openclaw/blob/main//docs/help/index.md | main | opebclaw | [
-0.015173775143921375,
-0.029153700917959213,
-0.020963845774531364,
0.0303980503231287,
0.10664994269609451,
-0.09068376570940018,
-0.007879345677793026,
0.0050578066147863865,
-0.08666238933801651,
0.0032547859009355307,
-0.013375923968851566,
-0.02783968485891819,
-0.054727546870708466,
... | 0.099823 |
# GPT-5.4 / Codex Parity Maintainer Notes This note explains how to review the GPT-5.4 / Codex parity program as four merge units without losing the original six-contract architecture. ## Merge units ### PR A: strict-agentic execution Owns: - `executionContract` - GPT-5-first same-turn follow-through - `update\_plan` a... | https://github.com/openclaw/openclaw/blob/main//docs/help/gpt54-codex-agentic-parity-maintainers.md | main | opebclaw | [
-0.07613841444253922,
-0.08670451492071152,
-0.06134554371237755,
-0.030446533113718033,
0.034879785031080246,
-0.08577347546815872,
-0.0015931347152218223,
0.0298185795545578,
-0.06657850742340088,
0.012795604765415192,
-0.016378186643123627,
0.03223714604973793,
-0.06466198712587357,
-0.... | 0.08222 |
H -- "yes" --> I["Parity claim allowed"] H -- "no" --> J["Keep runtime fixes / review loop open"] ``` The parity harness is not the only evidence source. Keep this split explicit in review: - PR D owns the scenario-based GPT-5.4 vs Opus 4.6 comparison - PR B deterministic suites still own auth/proxy/DNS and full-access... | https://github.com/openclaw/openclaw/blob/main//docs/help/gpt54-codex-agentic-parity-maintainers.md | main | opebclaw | [
-0.11740204691886902,
-0.043283190578222275,
-0.033743489533662796,
-0.0466463640332222,
0.05144503340125084,
-0.0012872179504483938,
-0.0171810369938612,
0.02710467018187046,
-0.025015538558363914,
0.032839372754096985,
0.016822300851345062,
0.031211482360959053,
-0.07276137918233871,
-0.... | 0.020767 |
# Debugging This page covers debugging helpers for streaming output, especially when a provider mixes reasoning into normal text. ## Runtime debug overrides Use `/debug` in chat to set \*\*runtime-only\*\* config overrides (memory, not disk). `/debug` is disabled by default; enable with `commands.debug: true`. This is ... | https://github.com/openclaw/openclaw/blob/main//docs/help/debugging.md | main | opebclaw | [
-0.04491815343499184,
-0.004608094226568937,
-0.017761435359716415,
0.08581168949604034,
0.048868365585803986,
-0.11061986535787582,
-0.016931042075157166,
0.032080963253974915,
-0.0227391067892313,
0.008905314840376377,
0.01893792860209942,
-0.024553028866648674,
-0.05726171284914017,
0.0... | 0.05544 |
`~/.openclaw/logs/raw-stream.jsonl` ## Raw chunk logging (pi-mono) To capture \*\*raw OpenAI-compat chunks\*\* before they are parsed into blocks, pi-mono exposes a separate logger: ```bash PI\_RAW\_STREAM=1 ``` Optional path: ```bash PI\_RAW\_STREAM\_PATH=~/.pi-mono/logs/raw-openai-completions.jsonl ``` Default file: ... | https://github.com/openclaw/openclaw/blob/main//docs/help/debugging.md | main | opebclaw | [
-0.056261833757162094,
0.004648149013519287,
0.005809897556900978,
0.011372017674148083,
0.06227710098028183,
-0.0852789655327797,
-0.03434330224990845,
0.02687729150056839,
0.0716940313577652,
0.013563952408730984,
-0.016472943127155304,
-0.07817642390727997,
-0.040307410061359406,
0.0712... | 0.109676 |
# QA Refactor Status: foundational migration landed. ## Goal Move OpenClaw QA from a split-definition model to a single source of truth: - scenario metadata - prompts sent to the model - setup and teardown - harness logic - assertions and success criteria - artifacts and report hints The desired end state is a generic ... | https://github.com/openclaw/openclaw/blob/main//docs/refactor/qa.md | main | opebclaw | [
-0.08643556386232376,
0.0031812521629035473,
-0.02418532967567444,
0.028155632317066193,
0.041293129324913025,
-0.004660763777792454,
-0.08500434458255768,
0.05053640156984329,
-0.020478159189224243,
0.07171100378036499,
-0.03250212222337723,
-0.0366983599960804,
-0.034256711602211,
0.0920... | 0.103974 |
[openai, qa-channel] docsRefs: - docs/help/testing.md - docs/concepts/model-providers.md codeRefs: - extensions/qa-lab/src/suite.ts - src/gateway/chat-attachments.ts --- # Objective Verify generated media is reattached on the follow-up turn. # Setup ```yaml scenario.setup - action: config.patch patch: agents: defaults:... | https://github.com/openclaw/openclaw/blob/main//docs/refactor/qa.md | main | opebclaw | [
-0.02056601643562317,
0.010408245027065277,
-0.01408194936811924,
0.052490610629320145,
0.051239609718322754,
-0.04693254455924034,
-0.020231494680047035,
-0.014818565919995308,
0.07713227719068527,
0.07577428221702576,
0.06494791060686111,
-0.05150562524795532,
-0.01340585295110941,
0.086... | 0.085931 |
Phase 2: generic engine - split `extensions/qa-lab/src/suite.ts` into: - loader - engine - action registry - assertion registry - custom handlers - keep existing helper functions as engine operations Deliverable: - engine executes simple declarative scenarios Start with scenarios that are mostly prompt + wait + assert:... | https://github.com/openclaw/openclaw/blob/main//docs/refactor/qa.md | main | opebclaw | [
-0.0695517286658287,
-0.00451234495267272,
0.03516948223114014,
0.058864954859018326,
0.013384802266955376,
0.037218235433101654,
-0.05607409402728081,
0.05933794006705284,
-0.10561618953943253,
0.02911420911550522,
0.003219673177227378,
-0.020468225702643394,
-0.02745828777551651,
0.03849... | 0.159395 |
migration | https://github.com/openclaw/openclaw/blob/main//docs/refactor/qa.md | main | opebclaw | [
0.034417349845170975,
-0.07621214538812637,
-0.009576567448675632,
-0.06390314549207687,
0.07924722880125046,
-0.007126928307116032,
0.011338358744978905,
-0.04820379242300987,
-0.08709707856178284,
-0.0008174012182280421,
0.04255890101194382,
-0.04197176918387413,
0.018701715394854546,
-0... | 0.061196 |
# Tailscale (Gateway dashboard) OpenClaw can auto-configure Tailscale \*\*Serve\*\* (tailnet) or \*\*Funnel\*\* (public) for the Gateway dashboard and WebSocket port. This keeps the Gateway bound to loopback while Tailscale provides HTTPS, routing, and (for Serve) identity headers. ## Modes - `serve`: Tailnet-only Serv... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/tailscale.md | main | opebclaw | [
-0.03145990148186684,
0.012129629962146282,
-0.04644620418548584,
0.03671389818191528,
-0.004700355231761932,
-0.08696397393941879,
-0.02112404815852642,
0.0303982961922884,
0.021239493042230606,
-0.04182792827486992,
-0.002709151478484273,
-0.006564595736563206,
0.016407255083322525,
-0.0... | 0.017227 |
+ limits - Serve requires HTTPS enabled for your tailnet; the CLI prompts if it is missing. - Serve injects Tailscale identity headers; Funnel does not. - Funnel requires Tailscale v1.38.3+, MagicDNS, HTTPS enabled, and a funnel node attribute. - Funnel only supports ports `443`, `8443`, and `10000` over TLS. - Funnel ... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/tailscale.md | main | opebclaw | [
0.015256655402481556,
0.0036524785682559013,
0.006939497776329517,
-0.06907215714454651,
-0.0405423529446125,
-0.1115606352686882,
-0.00923649687319994,
-0.005543439649045467,
0.0038198428228497505,
-0.03566848859190941,
-0.00788306538015604,
-0.02656034380197525,
-0.01374022662639618,
0.0... | -0.008086 |
# Authentication (Model Providers) This page covers \*\*model provider\*\* authentication (API keys, OAuth, Claude CLI reuse, and Anthropic setup-token). For \*\*gateway connection\*\* authentication (token, password, trusted-proxy), see [Configuration](/gateway/configuration) and [Trusted Proxy Auth](/gateway/trusted-... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/authentication.md | main | opebclaw | [
-0.08027396351099014,
-0.045480307191610336,
-0.09095003455877304,
-0.005253291688859463,
0.0018071280792355537,
-0.04164285212755203,
0.010087096132338047,
0.03978688642382622,
0.03289872407913208,
-0.034881338477134705,
0.014357181265950203,
0.011762620881199837,
0.05997041240334511,
0.0... | 0.07023 |
OpenClaw therefore treats Claude CLI reuse and `claude -p` usage as sanctioned for Anthropic-backed runs unless Anthropic publishes a new policy. - Anthropic API keys remain the most predictable choice for long-lived gateway hosts and explicit server-side billing control. ## Checking model auth status ```bash openclaw ... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/authentication.md | main | opebclaw | [
-0.1018873006105423,
0.005775493569672108,
-0.05592118203639984,
-0.018788857385516167,
-0.016610953956842422,
-0.09536663442850113,
-0.06158866733312607,
-0.03471536934375763,
-0.003833001246675849,
0.008589235134422779,
0.03480582684278488,
0.020202510058879852,
0.02925124391913414,
-0.0... | -0.009895 |
# Configuration Reference Core config reference for `~/.openclaw/openclaw.json`. For a task-oriented overview, see [Configuration](/gateway/configuration). This page covers the main OpenClaw config surfaces and links out when a subsystem has its own deeper reference. It does \*\*not\*\* try to inline every channel/plug... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.04971436411142349,
-0.054060470312833786,
-0.08958493173122406,
0.042335718870162964,
-0.0019426451763138175,
-0.08463756740093231,
-0.07681151479482651,
0.010931143537163734,
-0.064954474568367,
0.003351912135258317,
0.03694826364517212,
0.004882521461695433,
-0.045585982501506805,
0.0... | 0.063915 |
disabled allowFrom: ["+15555550123", "+447700900123"], textChunkLimit: 4000, chunkMode: "length", // length | newline mediaMaxMb: 50, sendReadReceipts: true, // blue ticks (false in self-chat mode) groups: { "\*": { requireMention: true }, }, groupPolicy: "allowlist", groupAllowFrom: ["+15551234567"], }, }, web: { enab... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.062181729823350906,
0.02261231653392315,
0.025263672694563866,
0.04001078009605408,
0.015736574307084084,
-0.055437251925468445,
-0.003088457742705941,
-0.06322857737541199,
0.006432363297790289,
0.016433333978056908,
0.020123668015003204,
-0.019564801827073097,
-0.07169733941555023,
0.... | 0.03075 |
24, maxAgeHours: 0, spawnSubagentSessions: false, // opt-in for sessions\_spawn({ thread: true }) }, voice: { enabled: true, autoJoin: [ { guildId: "123456789012345678", channelId: "234567890123456789", }, ], daveEncryption: true, decryptionFailureTolerance: 24, tts: { provider: "openai", openai: { voice: "alloy" }, },... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.058691155165433884,
0.01587136834859848,
-0.027472665533423424,
-0.037279870361089706,
-0.0006858600536361337,
-0.04869014397263527,
0.0910634845495224,
-0.06525585800409317,
0.04711933434009552,
0.027193548157811165,
0.02805432677268982,
-0.09209644049406052,
-0.04300478845834732,
0.05... | 0.125101 |
(from `guilds..users` on all messages). ### Google Chat ```json5 { channels: { googlechat: { enabled: true, serviceAccountFile: "/path/to/service-account.json", audienceType: "app-url", // app-url | project-number audience: "https://gateway.example.com/googlechat", webhookPath: "/googlechat", botUser: "users/1234567890... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.09552666544914246,
0.0072627984918653965,
0.027530964463949203,
0.008812439627945423,
-0.021829089149832726,
-0.056580692529678345,
0.05651058256626129,
-0.0435154102742672,
-0.014470584690570831,
0.03001004457473755,
0.016154654324054718,
-0.04077119007706642,
-0.010139121674001217,
0.... | -0.003244 |
| | messages | enabled | Read/send/edit/delete | | pins | enabled | Pin/unpin/list | | memberInfo | enabled | Member info | | emojiList | enabled | Custom emoji list | ### Mattermost Mattermost ships as a plugin: `openclaw plugins install @openclaw/mattermost`. ```json5 { channels: { mattermost: { enabled: true, botTok... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.06421644240617752,
-0.052333079278469086,
0.021977653726935387,
0.06904804706573486,
0.033328376710414886,
-0.0769512951374054,
-0.004473508335649967,
-0.041897960007190704,
0.04700927436351776,
-0.013335810042917728,
0.04146071895956993,
-0.043070659041404724,
-0.038391899317502975,
0.... | 0.11787 |
`/Users/\*/Library/Messages/Attachments`). - SCP uses strict host-key checking, so ensure the relay host key already exists in `~/.ssh/known\_hosts`. - `channels.imessage.configWrites`: allow or deny iMessage-initiated config writes. - Top-level `bindings[]` entries with `type: "acp"` can bind iMessage conversations to... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.0004711451183538884,
0.002116856398060918,
-0.010975352488458157,
0.05444749444723129,
0.02947928011417389,
-0.0222668144851923,
0.012088651768863201,
-0.029343612492084503,
0.028770947828888893,
0.03581135347485542,
0.025766199454665184,
0.04470972344279289,
-0.016912279650568962,
0.16... | 0.099668 |
add a non-default account via `openclaw channels add` (or channel onboarding) while still on a single-account top-level channel config, OpenClaw promotes account-scoped top-level single-account values into the channel account map first so the original account keeps working. Most channels move them into `channels..accou... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.001753367017954588,
-0.0923980250954628,
-0.07726658880710602,
0.00964883342385292,
-0.03699154034256935,
-0.04078112170100212,
0.01046188734471798,
-0.014615673571825027,
-0.01741458661854267,
-0.04331861808896065,
0.0367276631295681,
-0.020608296617865562,
0.009201949462294579,
0.0409... | 0.035491 |
`false` clears previously registered commands. - Override native skill registration per channel with `channels..commands.nativeSkills`. - `channels.telegram.customCommands` adds extra Telegram bot menu entries. - `bash: true` enables `! ` for host shell. Requires `tools.elevated.enabled` and sender in `tools.elevated.a... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.02082187309861183,
-0.09327314049005508,
-0.014353475533425808,
0.025033462792634964,
-0.03895283862948418,
-0.07747898995876312,
0.002696540905162692,
-0.007303646765649319,
-0.03005805052816868,
0.06454043090343475,
0.05682879686355591,
-0.0012577950255945325,
0.010168950073421001,
0.... | 0.173098 |
```json5 { agents: { defaults: { bootstrapPromptTruncationWarning: "once" } }, // off | once | always } ``` ### `agents.defaults.imageMaxDimensionPx` Max pixel size for the longest image side in transcript/tool image blocks before provider calls. Default: `1200`. Lower values usually reduce vision-token usage and reque... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.012101294472813606,
0.0665200725197792,
-0.00423564063385129,
-0.022697068750858307,
0.07701802998781204,
-0.04936937987804413,
-0.0747026726603508,
0.029655076563358307,
0.0035113368649035692,
0.019034693017601967,
0.008307474665343761,
-0.000938993995077908,
-0.053132008761167526,
0.1... | 0.072866 |
auth/API key too. - The bundled Qwen video-generation provider supports up to 1 output video, 1 input image, 4 input videos, 10 seconds duration, and provider-level `size`, `aspectRatio`, `resolution`, `audio`, and `watermark` options. - `pdfModel`: accepts either a string (`"provider/model"`) or an object (`{ primary,... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.02370510809123516,
-0.0852103903889656,
-0.16630826890468597,
-0.04431537166237831,
-0.02511064149439335,
0.04616100713610649,
-0.06952396780252457,
0.04108104854822159,
-0.011273126117885113,
0.015358028002083302,
0.051364682614803314,
-0.015364963561296463,
0.08458484709262848,
0.0764... | 0.013473 |
| `gemini` | `google/gemini-3.1-pro-preview` | | `gemini-flash` | `google/gemini-3-flash-preview` | | `gemini-flash-lite` | `google/gemini-3.1-flash-lite-preview` | Your configured aliases always win over defaults. Z.AI GLM-4.x models automatically enable thinking mode unless you set `--thinking off` or define `agents.... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.06435473263263702,
-0.06736946105957031,
-0.03456788882613182,
-0.0029273477848619223,
0.05477749928832054,
-0.06084388867020607,
-0.016722332686185837,
0.01300834957510233,
-0.03206845745444298,
-0.01198559906333685,
0.009709153324365616,
0.019040757790207863,
-0.007388888392597437,
0.... | 0.019882 |
compaction-only model override notifyUser: true, // send a brief notice when compaction starts (default: false) memoryFlush: { enabled: true, softThresholdTokens: 6000, systemPrompt: "Session nearing compaction. Store durable memories now.", prompt: "Write any lasting notes to memory/YYYY-MM-DD.md; reply with the exact... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.03429242968559265,
-0.017523910850286484,
-0.07235755026340485,
0.08172105252742767,
-0.007409371435642242,
0.03247417137026787,
-0.027964215725660324,
0.05980275571346283,
0.0534597709774971,
-0.010822528973221779,
0.04611773043870926,
0.04026567190885544,
0.03847388178110123,
0.014785... | 0.148265 |
direct chats/mentions, `message` for unmentioned group chats. - Per-session overrides: `session.typingMode`, `session.typingIntervalSeconds`. See [Typing Indicators](/concepts/typing-indicators). ### `agents.defaults.sandbox` Optional sandboxing for the embedded agent. See [Sandboxing](/gateway/sandboxing) for the full... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
0.0006660472135990858,
0.013513494282960892,
-0.002525999676436186,
0.04361547529697418,
0.07024461030960083,
-0.028113501146435738,
0.033784471452236176,
-0.05641373619437218,
0.05574819818139076,
-0.04213111102581024,
0.026929227635264397,
-0.038649413734674454,
-0.034552935510873795,
0.... | 0.157709 |
- `mirror`: seed remote from local before exec, sync back after exec; local workspace stays canonical - `remote`: seed remote once when the sandbox is created, then keep the remote workspace canonical In `remote` mode, host-local edits made outside OpenClaw are not synced into the sandbox automatically after the seed s... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
0.08278490602970123,
-0.022477321326732635,
-0.02174021489918232,
0.026571357622742653,
0.03507761284708977,
-0.01825726218521595,
-0.09176936745643616,
-0.07328620553016663,
-0.016921397298574448,
0.055453017354011536,
0.020890770480036736,
-0.06757716089487076,
0.045644741505384445,
-0.0... | -0.019877 |
wins (warning logged). If none set, first list entry is default. - `model`: string form overrides `primary` only; object form `{ primary, fallbacks }` overrides both (`[]` disables global fallbacks). Cron jobs that only override `primary` still inherit default fallbacks unless you set `fallbacks: []`. - `params`: per-a... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.04896269366145134,
-0.06013735011219978,
-0.0697648674249649,
0.03943498060107231,
-0.035596199333667755,
0.031057562679052353,
-0.0057024601846933365,
0.04025546833872795,
-0.1258910745382309,
-0.015828164294362068,
0.016656450927257538,
-0.021986767649650574,
0.030206220224499702,
0.0... | 0.059092 |
list: [ { id: "public", workspace: "~/.openclaw/workspace-public", sandbox: { mode: "all", scope: "agent", workspaceAccess: "none" }, tools: { allow: [ "sessions\_list", "sessions\_history", "sessions\_send", "sessions\_spawn", "session\_status", "whatsapp", "telegram", "slack", "discord", "gateway", ], deny: [ "read",... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.04298083111643791,
0.07634183019399643,
-0.06126970797777176,
0.027244677767157555,
0.05530236288905144,
-0.05590466409921646,
0.04476478695869446,
-0.027372093871235847,
-0.021055445075035095,
0.04843369871377945,
0.00432704109698534,
-0.036650918424129486,
-0.002009877236559987,
0.070... | 0.14549 |
- \*\*`threadBindings`\*\*: global defaults for thread-bound session features. - `enabled`: master default switch (providers can override; Discord uses `channels.discord.threadBindings.enabled`) - `idleHours`: default inactivity auto-unfocus in hours (`0` disables; providers can override) - `maxAgeHours`: default hard ... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.06971181929111481,
-0.010641802102327347,
0.03045046143233776,
0.07454545050859451,
-0.011164595372974873,
-0.000031227009458234534,
0.09562292695045471,
-0.053435541689395905,
0.059774503111839294,
0.0030360128730535507,
0.04245053976774216,
-0.03518887609243393,
-0.052003998309373856,
... | 0.140618 |
key in `talk.providers` when multiple Talk providers are configured. - Legacy flat Talk keys (`talk.voiceId`, `talk.voiceAliases`, `talk.modelId`, `talk.outputFormat`, `talk.apiKey`) are compatibility-only and are auto-migrated into `talk.providers.`. - Voice IDs fall back to `ELEVENLABS\_VOICE\_ID` or `SAG\_VOICE\_ID`... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.09321112185716629,
-0.0587502047419548,
-0.026041073724627495,
-0.05200369656085968,
-0.033575836569070816,
-0.008296159096062183,
0.007943694479763508,
-0.03871503472328186,
-0.004098952282220125,
-0.030372507870197296,
0.017779206857085228,
0.0043917689472436905,
-0.006505562923848629,
... | 0.062592 |
warn/block on known poll tools (`process.poll`, `command\_status`, etc.). - `detectors.pingPong`: warn/block on alternating no-progress pair patterns. - If `warningThreshold >= criticalThreshold` or `criticalThreshold >= globalCircuitBreakerThreshold`, validation fails. ### `tools.web` ```json5 { tools: { web: { search... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.07344304025173187,
0.039806216955184937,
-0.02941678836941719,
0.03181840851902962,
0.04582883417606354,
-0.08859884738922119,
-0.05185266211628914,
-0.0573003888130188,
-0.00676985178142786,
0.05787346884608269,
-0.01602872833609581,
-0.06501619517803192,
-0.029168201610445976,
0.00492... | 0.049362 |
removes attachments; `keep` retains them only when `retainOnSessionKeep: true`. ### `tools.experimental` Experimental built-in tool flags. Default off unless a strict-agentic GPT-5 auto-enable rule applies. ```json5 { tools: { experimental: { planTool: true, // enable experimental update\_plan }, }, } ``` Notes: - `pla... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.04963091015815735,
0.02847583405673504,
-0.023386908695101738,
0.1134491041302681,
0.11301184445619583,
-0.08050977438688278,
-0.004228533711284399,
-0.043141257017850876,
0.044959232211112976,
-0.006882421672344208,
0.06894446909427643,
0.07561992853879929,
-0.04712752625346184,
0.0519... | 0.128821 |
URL. - `models.providers.\*.headers`: extra static headers for proxy/tenant routing. - `models.providers.\*.request`: transport overrides for model-provider HTTP requests. - `request.headers`: extra headers (merged with provider defaults). Values accept SecretRef. - `request.auth`: auth strategy override. Modes: `"prov... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.07596399635076523,
0.02692979760468006,
-0.018180858343839645,
-0.010296436958014965,
-0.03953895345330238,
-0.049705564975738525,
0.013084057718515396,
-0.00020113457867410034,
0.008727433159947395,
0.045839373022317886,
-0.0007896625902503729,
-0.011615262366831303,
0.020780479535460472... | -0.008286 |
Native Moonshot endpoints advertise streaming usage compatibility on the shared `openai-completions` transport, and OpenClaw keys that off endpoint capabilities rather than the built-in provider id alone. ```json5 { env: { KIMI\_API\_KEY: "sk-..." }, agents: { defaults: { model: { primary: "kimi/kimi-code" }, models: {... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.0006573442369699478,
-0.04198114946484566,
-0.029605945572257042,
-0.002738227602094412,
0.062040746212005615,
-0.030976682901382446,
-0.04927660897374153,
-0.05114592984318733,
-0.0069680786691606045,
-0.020345108583569527,
0.044335756450891495,
-0.05280182510614395,
-0.01918422617018222... | 0.082702 |
and `providerOverride`. Applies to native plugin hooks and supported bundle-provided hook directories. - `plugins.entries..subagent.allowModelOverride`: explicitly trust this plugin to request per-run `provider` and `model` overrides for background subagent runs. - `plugins.entries..subagent.allowedModels`: optional al... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.05479671433568001,
-0.04316150024533272,
-0.00782979466021061,
0.03922094404697418,
0.03625280782580376,
0.003173680044710636,
-0.055321384221315384,
-0.010998665355145931,
-0.09765738248825073,
-0.017274683341383934,
0.05150419473648071,
-0.013095926493406296,
-0.023080237209796906,
0.... | 0.025 |
target a specific Chromium-based browser profile such as Brave or Edge. - `existing-session` profiles keep the current Chrome MCP route limits: snapshot/ref-driven actions instead of CSS-selector targeting, one-file upload hooks, no dialog timeout overrides, no `wait --load networkidle`, and no `responsebody`, PDF expo... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.04145871847867966,
0.022643841803073883,
0.026283225044608116,
0.05425775796175003,
0.05274799093604088,
-0.03167622163891792,
-0.11594070494174957,
0.05958348140120506,
-0.10589195787906647,
-0.060017552226781845,
0.01808885857462883,
-0.028162848204374313,
-0.031993091106414795,
0.002... | -0.043337 |
not offered by onboarding prompts. - `gateway.auth.mode: "trusted-proxy"`: delegate auth to an identity-aware reverse proxy and trust identity headers from `gateway.trustedProxies` (see [Trusted Proxy Auth](/gateway/trusted-proxy-auth)). This mode expects a \*\*non-loopback\*\* proxy source; same-host loopback reverse ... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.0593746043741703,
0.01589907705783844,
0.006567873526364565,
-0.00957550760358572,
-0.003730451688170433,
-0.04584217071533203,
0.06068995222449303,
0.04480423405766487,
0.025884583592414856,
0.02095215395092964,
-0.013133827596902847,
-0.02353012003004551,
0.0694630891084671,
0.0128587... | 0.018639 |
`trustedProxies`: reverse proxy IPs that terminate TLS or inject forwarded-client headers. Only list proxies you control. Loopback entries are still valid for same-host proxy/local-detection setups (for example Tailscale Serve or a local reverse proxy), but they do \*\*not\*\* make loopback requests eligible for `gatew... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.050358157604932785,
-0.00903367530554533,
-0.008843788877129555,
-0.0015333091141656041,
0.004636061377823353,
-0.07449498772621155,
0.003910750150680542,
-0.048359084874391556,
0.029291929677128792,
0.037505730986595154,
0.04518177732825279,
0.045930370688438416,
-0.02059289440512657,
... | 0.01615 |
`/hooks/gmail` → `gmail`). - `match.source` matches a payload field for generic paths. - Templates like `{{messages[0].subject}}` read from the payload. - `transform` can point to a JS/TS module returning a hook action. - `transform.module` must be a relative path and stays within `hooks.transformsDir` (absolute paths ... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.11374654620885849,
0.04931418597698212,
0.034234847873449326,
0.003798894817009568,
-0.008773310109972954,
-0.018860602751374245,
0.04304460063576698,
-0.010779788717627525,
0.01142012607306242,
0.025066057220101357,
-0.017975924536585808,
-0.002416372997686267,
-0.019932013005018234,
0... | 0.106364 |
throw an error at config load. - Escape with `$${VAR}` for a literal `${VAR}`. - Works with `$include`. --- ## Secrets Secret refs are additive: plaintext values still work. ### `SecretRef` Use one object shape: ```json5 { source: "env" | "file" | "exec", provider: "default", id: "..." } ``` Validation: - `provider` pa... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.10973893851041794,
0.06415732204914093,
-0.005696114618331194,
0.027001002803444862,
0.04893568530678749,
-0.03619414567947388,
0.018863575533032417,
-0.017216354608535767,
0.04751810431480408,
0.0020371454302221537,
0.04986755549907684,
-0.044637635350227356,
0.01202867366373539,
0.073... | 0.046878 |
window in hours used for backoff counters (default: `24`). - `overloadedProfileRotations`: maximum same-provider auth-profile rotations for overloaded errors before switching to model fallback (default: `1`). Provider-busy shapes such as `ModelNotReadyException` land here. - `overloadedBackoffMs`: fixed delay before re... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.056095246225595474,
-0.005877215880900621,
-0.00025001526228152215,
0.08964461833238602,
0.04909130185842514,
-0.07825659215450287,
-0.07198156416416168,
0.010368779301643372,
-0.00594449695199728,
0.0008506028680130839,
0.004180096089839935,
-0.0007898097392171621,
-0.01863533817231655,
... | 0.030355 |
`false` to keep ACP commands available while blocking execution. - `backend`: default ACP runtime backend id (must match a registered ACP runtime plugin). - `defaultAgent`: fallback ACP target agent id when spawns do not specify an explicit target. - `allowedAgents`: allowlist of agent ids permitted for ACP runtime ses... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.08482576906681061,
0.015406496822834015,
-0.10881561785936356,
0.02347104623913765,
0.025062980130314827,
0.006002351641654968,
-0.01651027984917164,
0.0007130292942747474,
-0.025953643023967743,
0.036554086953401566,
0.03977304697036743,
-0.04904543235898018,
-0.03594283014535904,
-0.0... | 0.0611 |
`"timeout"`, `"server\_error"`. Omit to retry all transient types. Applies only to one-shot cron jobs. Recurring jobs use separate failure handling. ### `cron.failureAlert` ```json5 { cron: { failureAlert: { enabled: false, after: 3, cooldownMs: 3600000, mode: "announce", accountId: "main", }, }, } ``` - `enabled`: ena... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/configuration-reference.md | main | opebclaw | [
-0.08004330098628998,
0.04237290471792221,
0.00917869247496128,
0.06964553147554398,
0.0371178537607193,
-0.08051405847072601,
-0.007583355065435171,
-0.03753456473350525,
0.005877492483705282,
-0.02218473330140114,
0.014076928608119488,
-0.02889300510287285,
-0.013053291477262974,
0.01647... | 0.064992 |
# CLI backends (fallback runtime) OpenClaw can run \*\*local AI CLIs\*\* as a \*\*text-only fallback\*\* when API providers are down, rate-limited, or temporarily misbehaving. This is intentionally conservative: - \*\*OpenClaw tools are not injected directly\*\*, but backends with `bundleMcp: true` can receive gateway ... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/cli-backends.md | main | opebclaw | [
-0.10836682468652725,
0.007215963676571846,
-0.015820477157831192,
0.02662663720548153,
0.040615200996398926,
-0.08552148938179016,
-0.005811664741486311,
0.02708679623901844,
0.04664643481373787,
-0.04371799901127815,
0.014884324744343758,
-0.05340653657913208,
0.04775531589984894,
0.0543... | 0.173544 |
config override (`-c model\_instructions\_file="..."`). Codex does not expose a Claude-style `--append-system-prompt` flag, so OpenClaw writes the assembled prompt to a temporary file for each fresh Codex CLI session. The bundled Anthropic `claude-cli` backend receives the OpenClaw skills snapshot two ways: the compact... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/cli-backends.md | main | opebclaw | [
-0.05393921956419945,
0.02140827663242817,
-0.11361073702573776,
0.018900999799370766,
-0.026056919246912003,
-0.061755213886499405,
-0.008839337155222893,
0.03335673734545708,
-0.04952900484204292,
-0.016192419454455376,
-0.011241324245929718,
-0.006274896208196878,
0.019413325935602188,
... | -0.014367 |
absolute `command` path). ## Plugin-owned defaults CLI backend defaults are now part of the plugin surface: - Plugins register them with `api.registerCliBackend(...)`. - The backend `id` becomes the provider prefix in model refs. - User config in `agents.defaults.cliBackends.` still overrides the plugin default. - Back... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/cli-backends.md | main | opebclaw | [
-0.07494302093982697,
-0.030272426083683968,
0.007648029364645481,
0.04287434741854668,
-0.10497264564037323,
0.07938848435878754,
0.03335915133357048,
0.07108105719089508,
0.04141698405146599,
-0.0192713662981987,
-0.010794080793857574,
0.0029428491834551096,
-0.036953046917915344,
0.1214... | 0.095181 |
# Bonjour / mDNS discovery OpenClaw uses Bonjour (mDNS / DNS‑SD) to discover an active Gateway (WebSocket endpoint). Multicast `local.` browsing is a \*\*LAN-only convenience\*\*. For cross-network discovery, the same beacon can also be published through a configured wide-area DNS-SD domain. Discovery is still best-eff... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/bonjour.md | main | opebclaw | [
-0.029171345755457878,
-0.03933300822973251,
0.013365655206143856,
-0.04792245477437973,
0.002079309429973364,
-0.08423623442649841,
-0.05672813206911087,
-0.043302007019519806,
0.00660654716193676,
0.0004404233186505735,
-0.026699917390942574,
-0.00806677807122469,
0.011611595749855042,
-... | -0.010823 |
dns-sd -B \_openclaw-gw.\_tcp local. ``` - Resolve one instance (replace ``): ```bash dns-sd -L "" \_openclaw-gw.\_tcp local. ``` If browsing works but resolving fails, you’re usually hitting a LAN policy or mDNS resolver issue. ## Debugging in Gateway logs The Gateway writes a rolling log file (printed on startup as `... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/bonjour.md | main | opebclaw | [
-0.04304351657629013,
0.022645076736807823,
0.06863237172365189,
-0.026074476540088654,
-0.05265675485134125,
-0.10308503359556198,
-0.08753739297389984,
-0.0370744951069355,
0.018654774874448776,
0.04830571264028549,
-0.04014792665839195,
-0.02864510379731655,
-0.05624215304851532,
0.0405... | -0.009165 |
# Local models Local is doable, but OpenClaw expects large context + strong defenses against prompt injection. Small cards truncate context and leak safety. Aim high: \*\*≥2 maxed-out Mac Studios or equivalent GPU rig (~$30k+)\*\*. A single \*\*24 GB\*\* GPU works only for lighter prompts with higher latency. Use the \... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/local-models.md | main | opebclaw | [
0.04921148344874382,
-0.0037067746743559837,
0.007214522920548916,
0.036324746906757355,
0.035235337913036346,
-0.09867335855960846,
-0.14067873358726501,
0.035410456359386444,
-0.017106562852859497,
-0.018456313759088516,
0.014905398711562157,
-0.03502245247364044,
0.03162884712219238,
-0... | 0.035959 |
{ input: 0, output: 0, cacheRead: 0, cacheWrite: 0 }, contextWindow: 120000, maxTokens: 8192, }, ], }, }, }, } ``` Keep `models.mode: "merge"` so hosted models stay available as fallbacks. Behavior note for local/proxied `/v1` backends: - OpenClaw treats these as proxy-style OpenAI-compatible routes, not native OpenAI ... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/local-models.md | main | opebclaw | [
-0.024653293192386627,
-0.05660649389028549,
-0.012068967334926128,
0.06813524663448334,
0.028256328776478767,
-0.04732620716094971,
-0.04674029350280762,
-0.02171977609395981,
0.00920896790921688,
0.0065986174158751965,
-0.0008545445743948221,
0.011245254427194595,
0.0014981577405706048,
... | 0.087678 |
# Secrets management OpenClaw supports additive SecretRefs so supported credentials do not need to be stored as plaintext in configuration. Plaintext still works. SecretRefs are opt-in per credential. ## Goals and runtime model Secrets are resolved into an in-memory runtime snapshot. - Resolution is eager during activa... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/secrets.md | main | opebclaw | [
-0.10586607456207275,
-0.029776886105537415,
-0.00652090972289443,
0.05704021826386452,
0.026372723281383514,
-0.04352833703160286,
-0.022670505568385124,
0.01562669314444065,
0.010391626507043839,
0.033526480197906494,
0.025655897334218025,
0.05725165456533432,
0.03897826746106148,
-0.029... | 0.069599 |
Use one object shape everywhere: ```json5 { source: "env" | "file" | "exec", provider: "default", id: "..." } ``` ### `source: "env"` ```json5 { source: "env", provider: "default", id: "OPENAI\_API\_KEY" } ``` Validation: - `provider` must match `^[a-z][a-z0-9\_-]{0,63}$` - `id` must match `^[A-Z][A-Z0-9\_]{0,127}$` ##... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/secrets.md | main | opebclaw | [
-0.07670526206493378,
0.08215886354446411,
0.002768973121419549,
-0.03292103111743927,
0.029867738485336304,
-0.06314141303300858,
0.009488830342888832,
0.005274263210594654,
-0.00490570580586791,
-0.008093378506600857,
0.028393466025590897,
-0.0667218416929245,
0.018036700785160065,
0.106... | 0.066949 |
id: "value" }, }, }, }, } ``` ### `sops` ```json5 { secrets: { providers: { sops\_openai: { source: "exec", command: "/opt/homebrew/bin/sops", allowSymlinkCommand: true, // required for Homebrew symlinked binaries trustedDirs: ["/opt/homebrew"], args: ["-d", "--extract", '["providers"]["openai"]["apiKey"]', "/path/to/s... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/secrets.md | main | opebclaw | [
-0.020935604348778725,
0.0517352931201458,
-0.049013901501894,
0.019018331542611122,
0.08314430713653564,
0.016647325828671455,
-0.028735902160406113,
0.02053864672780037,
0.006077243015170097,
-0.03714847192168236,
0.0367153137922287,
-0.02819538675248623,
-0.048077404499053955,
0.0563204... | 0.015914 |
a healthy state, OpenClaw enters degraded secrets state. One-shot system event and log codes: - `SECRETS\_RELOADER\_DEGRADED` - `SECRETS\_RELOADER\_RECOVERED` Behavior: - Degraded: runtime keeps last-known-good snapshot. - Recovered: emitted once after the next successful activation. - Repeated failures while already d... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/secrets.md | main | opebclaw | [
-0.03089539334177971,
-0.04885563254356384,
-0.05378267914056778,
0.09119859337806702,
0.03368847817182541,
-0.06621124595403671,
-0.10242880880832672,
0.02157079614698887,
0.009901636280119419,
0.005709543824195862,
0.016092045232653618,
0.06126773729920387,
0.02047698386013508,
0.0199987... | 0.057527 |
--allow-exec ``` Exec note: - dry-run skips exec checks unless `--allow-exec` is set. - write mode rejects plans containing exec SecretRefs/providers unless `--allow-exec` is set. For strict target/path contract details and exact rejection rules, see: - [Secrets Apply Plan Contract](/gateway/secrets-plan-contract) ## O... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/secrets.md | main | opebclaw | [
-0.09258413314819336,
0.036939430981874466,
-0.05769980698823929,
0.02009601518511772,
0.014337373897433281,
-0.02573140151798725,
-0.05346183106303215,
0.03966684266924858,
-0.044804494827985764,
0.03677993267774582,
0.03416750580072403,
0.007880741730332375,
0.005925508216023445,
-0.0187... | 0.089253 |
# OpenShell OpenShell is a managed sandbox backend for OpenClaw. Instead of running Docker containers locally, OpenClaw delegates sandbox lifecycle to the `openshell` CLI, which provisions remote environments with SSH-based command execution. The OpenShell plugin reuses the same core SSH transport and remote filesystem... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/openshell.md | main | opebclaw | [
0.07813608646392822,
0.009704319760203362,
-0.03384534642100334,
0.05641240254044533,
0.06303644180297852,
0.006757082883268595,
-0.14291301369667053,
-0.04907975718379021,
-0.0041870917193591595,
0.012149512767791748,
-0.00524816382676363,
-0.03651556745171547,
0.0072203222662210464,
0.04... | 0.072918 |
`"remote"` | `"mirror"` | Workspace sync mode | | `command` | `string` | `"openshell"` | Path or name of the `openshell` CLI | | `from` | `string` | `"openclaw"` | Sandbox source for first-time create | | `gateway` | `string` | — | OpenShell gateway name (`--gateway`) | | `gatewayEndpoint` | `string` | — | OpenShell ga... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/openshell.md | main | opebclaw | [
0.01919875293970108,
-0.08850616961717606,
-0.09233230352401733,
0.01825631409883499,
-0.015375147573649883,
0.011442153714597225,
-0.0032672130037099123,
-0.005959901958703995,
-0.04308313876390457,
0.02554461546242237,
0.028876004740595818,
-0.05006984621286392,
0.03496803343296051,
-0.0... | 0.049281 |
5. In `remote` mode: seed once on create, then operate directly on the remote workspace. ## See also - [Sandboxing](/gateway/sandboxing) -- modes, scopes, and backend comparison - [Sandbox vs Tool Policy vs Elevated](/gateway/sandbox-vs-tool-policy-vs-elevated) -- debugging blocked tools - [Multi-Agent Sandbox and Tool... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/openshell.md | main | opebclaw | [
0.0271353330463171,
-0.0735723152756691,
-0.05370083078742027,
0.00374743458814919,
0.051528915762901306,
-0.02751960977911949,
-0.076268270611763,
-0.05310845002532005,
-0.02880588173866272,
0.014010105282068253,
0.06713796406984329,
-0.036082614213228226,
0.04057628661394119,
-0.02372606... | 0.037623 |
# OpenAI Chat Completions (HTTP) OpenClaw’s Gateway can serve a small OpenAI-compatible Chat Completions endpoint. This endpoint is \*\*disabled by default\*\*. Enable it in config first. - `POST /v1/chat/completions` - Same port as the Gateway (WS + HTTP multiplex): `http://:/v1/chat/completions` When the Gateway’s Op... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/openai-http-api.md | main | opebclaw | [
-0.05198650807142258,
-0.037860043346881866,
-0.05719098448753357,
0.05769426375627518,
0.009370523504912853,
-0.10216082632541656,
-0.03394022211432457,
-0.05305028706789017,
0.04569163918495178,
-0.01984521746635437,
0.024742135778069496,
-0.0030852858908474445,
-0.04362151771783829,
0.0... | 0.089268 |
sets the synthetic ingress channel context for channel-aware prompts and policies. Compatibility aliases still accepted: - `model: "openclaw:"` - `model: "agent:"` ## Enabling the endpoint Set `gateway.http.endpoints.chatCompletions.enabled` to `true`: ```json5 { gateway: { http: { endpoints: { chatCompletions: { enabl... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/openai-http-api.md | main | opebclaw | [
-0.07014195621013641,
-0.01825619861483574,
-0.016247622668743134,
0.011893350630998611,
-0.015850108116865158,
-0.07837928831577301,
0.04512771964073181,
-0.03181299939751625,
0.08271391689777374,
0.03736073896288872,
-0.009269237518310547,
-0.01466396264731884,
-0.07007813453674316,
0.08... | 0.187736 |
`openclaw/default` is always present so one stable id works across environments. - Backend provider/model overrides belong in `x-openclaw-model`, not the OpenAI `model` field. - `/v1/embeddings` supports `input` as a string or array of strings. | https://github.com/openclaw/openclaw/blob/main//docs/gateway/openai-http-api.md | main | opebclaw | [
-0.027225686237215996,
-0.04582918435335159,
-0.07673182338476181,
0.056223053485155106,
0.060940735042095184,
-0.028829243034124374,
-0.07497923821210861,
-0.043434035032987595,
0.003706942079588771,
-0.08253119885921478,
0.05423596873879433,
-0.010600902140140533,
0.03927241265773773,
0.... | 0.057886 |
# Gateway troubleshooting This page is the deep runbook. Start at [/help/troubleshooting](/help/troubleshooting) if you want the fast triage flow first. ## Command ladder Run these first, in this order: ```bash openclaw status openclaw gateway status openclaw logs --follow openclaw doctor openclaw channels status --pro... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/troubleshooting.md | main | opebclaw | [
-0.015795782208442688,
-0.017268581315875053,
-0.10137703269720078,
0.024775447323918343,
0.006804953329265118,
-0.10521745681762695,
-0.04419601336121559,
-0.034262340515851974,
-0.08685901761054993,
0.007676201406866312,
0.03459510579705238,
-0.014542731456458569,
-0.06348969787359238,
-... | 0.157943 |
required` → group message ignored until mention. - `pairing request` → sender needs approval. - `blocked` / `allowlist` → sender/channel was filtered by policy. Related: - [/channels/troubleshooting](/channels/troubleshooting) - [/channels/pairing](/channels/pairing) - [/channels/groups](/channels/groups) ## Dashboard ... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/troubleshooting.md | main | opebclaw | [
-0.05773822218179703,
0.014301919378340244,
-0.10097117722034454,
0.030313696712255478,
0.00870419293642044,
-0.078171007335186,
-0.04771853983402252,
-0.03907667472958565,
-0.022358104586601257,
-0.04085826873779297,
0.07555323839187622,
-0.03321616351604462,
0.021027887240052223,
0.02604... | 0.012714 |
- paired-device token sessions can manage only \*\*their own\*\* device unless the caller also has `operator.admin` - `openclaw devices rotate --scope ...` can only request operator scopes that the caller session already holds Related: - [/web/control-ui](/web/control-ui) - [/gateway/configuration](/gateway/configurati... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/troubleshooting.md | main | opebclaw | [
-0.07535519450902939,
-0.001721343258395791,
-0.06926052272319794,
-0.018141888082027435,
-0.054913926869630814,
-0.09558017551898956,
0.050590697675943375,
-0.013201008550822735,
0.0037178928032517433,
0.0011537739774212241,
0.01626584492623806,
0.03885326161980629,
0.01544391643255949,
0... | 0.070027 |
```bash openclaw cron status openclaw cron list openclaw cron runs --id --limit 20 openclaw system heartbeat last openclaw logs --follow ``` Look for: - Cron enabled and next wake present. - Job run history status (`ok`, `skipped`, `error`). - Heartbeat skip reasons (`quiet-hours`, `requests-in-flight`, `alerts-disable... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/troubleshooting.md | main | opebclaw | [
-0.05118855461478233,
0.019760703667998314,
-0.06828623265028,
0.017448879778385162,
0.09445740282535553,
-0.08256988227367401,
-0.06788884103298187,
-0.06063804030418396,
-0.0003939236921723932,
-0.08971881121397018,
0.009461352601647377,
-0.04497359320521355,
0.004801837261766195,
0.0127... | 0.06117 |
stay unavailable. - `fullPage is not supported for element screenshots` → screenshot request mixed `--full-page` with `--ref` or `--element`. - `element screenshots are not supported for existing-session profiles; use ref from snapshot.` → Chrome MCP / `existing-session` screenshot calls must use page capture or a snap... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/troubleshooting.md | main | opebclaw | [
-0.09641154855489731,
0.012835019268095493,
0.07112520188093185,
0.03223484382033348,
0.12673719227313995,
-0.03049982152879238,
-0.03049805760383606,
0.023597760125994682,
-0.029855411499738693,
0.011859356425702572,
0.021373381838202477,
-0.004772626329213381,
0.018388492986559868,
-0.01... | 0.003083 |
# Doctor `openclaw doctor` is the repair + migration tool for OpenClaw. It fixes stale config/state, checks health, and provides actionable repair steps. ## Quick start ```bash openclaw doctor ``` ### Headless / automation ```bash openclaw doctor --yes ``` Accept defaults without prompting (including restart/service/sa... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/doctor.md | main | opebclaw | [
-0.0041515640914440155,
-0.006944926455616951,
-0.026855748146772385,
-0.014736275188624859,
0.03654634952545166,
-0.0570903941988945,
-0.02210378646850586,
0.01668095402419567,
-0.11463934183120728,
-0.020464560016989708,
-0.012779561802744865,
0.023411761969327927,
0.010796678252518177,
... | 0.115633 |
REM diary pass, and writes reversible backfill entries into `DREAMS.md`. - \*\*Reset\*\* removes only those marked backfill diary entries from `DREAMS.md`. - \*\*Clear Grounded\*\* removes only staged grounded-only short-term entries that came from historical replay and have not accumulated live recall or daily support... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/doctor.md | main | opebclaw | [
0.010144397616386414,
0.006449692416936159,
-0.041278399527072906,
-0.0020738481543958187,
-0.0036708619445562363,
0.0055839489214122295,
-0.02313600480556488,
-0.03013712354004383,
0.014546855352818966,
0.008865972980856895,
-0.024694113060832024,
0.07779830694198608,
0.04906314238905907,
... | 0.021557 |
API routing + costs. ### 2c) Browser migration and Chrome MCP readiness If your browser config still points at the removed Chrome extension path, doctor normalizes it to the current host-local Chrome MCP attach model: - `browser.profiles.\*.driver: "extension"` becomes `"existing-session"` - `browser.relayBindHost` is ... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/doctor.md | main | opebclaw | [
-0.021349070593714714,
-0.022909890860319138,
0.08744412660598755,
0.0229809433221817,
0.0786818265914917,
-0.01925557665526867,
-0.10004191845655441,
0.0690697506070137,
-0.08823603391647339,
-0.04183676466345787,
0.026581063866615295,
-0.02273988351225853,
0.007208165712654591,
0.0290545... | -0.011558 |
store migrations Doctor also checks the cron job store (`~/.openclaw/cron/jobs.json` by default, or `cron.store` when overridden) for old job shapes that the scheduler still accepts for compatibility. Current cron cleanups include: - `jobId` → `id` - `schedule.cron` → `schedule.expr` - top-level payload fields (`messag... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/doctor.md | main | opebclaw | [
-0.11266334354877472,
-0.013760560192167759,
-0.02531144581735134,
-0.008021065033972263,
0.04189682751893997,
-0.0948452427983284,
-0.04603118449449539,
-0.05425524339079857,
-0.05796508491039276,
-0.051820866763591766,
0.034869179129600525,
-0.02107074484229088,
-0.016666656360030174,
0.... | 0.082143 |
- longer disables (billing/credit failures) ### 6) Hooks model validation If `hooks.gmail.model` is set, doctor validates the model reference against the catalog and allowlist and warns when it won’t resolve or is disallowed. ### 7) Sandbox image repair When sandboxing is enabled, doctor checks Docker images and offers... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/doctor.md | main | opebclaw | [
-0.07980573177337646,
0.0077602253295481205,
0.07736334204673767,
-0.03976663947105408,
0.04961702227592468,
-0.04143401235342026,
-0.03684926778078079,
0.03138572722673416,
-0.08600767701864243,
-0.028471484780311584,
0.038794659078121185,
0.045925550162792206,
-0.026613203808665276,
0.00... | 0.090065 |
token and no token source exists, doctor offers to generate one. - If `gateway.auth.token` is SecretRef-managed but unavailable, doctor warns and does not overwrite it with plaintext. - `openclaw doctor --generate-gateway-token` forces generation only when no token SecretRef is configured. ### 12b) Read-only SecretRef-... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/doctor.md | main | opebclaw | [
-0.09407196938991547,
-0.00401442451402545,
-0.0026588046457618475,
0.025495795533061028,
-0.0017368054250255227,
-0.06774842739105225,
-0.02698858268558979,
0.07191900908946991,
0.037337254732847214,
0.03902256861329079,
-0.003358806250616908,
-0.00613787304610014,
0.017252430319786072,
-... | 0.118274 |
for port collisions on the gateway port (default `18789`) and reports likely causes (gateway already running, SSH tunnel). ### 17) Gateway runtime best practices Doctor warns when the gateway service runs on Bun or a version-managed Node path (`nvm`, `fnm`, `volta`, `asdf`, etc.). WhatsApp + Telegram channels require N... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/doctor.md | main | opebclaw | [
0.005676307715475559,
-0.028074011206626892,
0.013287820853292942,
-0.049748361110687256,
-0.007434500847011805,
-0.06282287836074829,
-0.05938778445124626,
0.08040667325258255,
-0.011295060627162457,
0.07773005217313766,
0.01839226670563221,
0.10412093997001648,
-0.02296794019639492,
0.03... | 0.051318 |
# Secrets apply plan contract This page defines the strict contract enforced by `openclaw secrets apply`. If a target does not match these rules, apply fails before mutating configuration. ## Plan file shape `openclaw secrets apply --from ` expects a `targets` array of plan targets: ```json5 { version: 1, protocolVersi... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/secrets-plan-contract.md | main | opebclaw | [
-0.040436144918203354,
0.060905274003744125,
0.011776834726333618,
0.012310350313782692,
0.07185239344835281,
-0.05198988318443298,
-0.040271908044815063,
-0.011999789625406265,
0.0048706429079174995,
-0.00226941448636353,
0.03574890270829201,
-0.020457729697227478,
0.017367642372846603,
0... | 0.026121 |
# Sandboxing OpenClaw can run \*\*tools inside sandbox backends\*\* to reduce blast radius. This is \*\*optional\*\* and controlled by configuration (`agents.defaults.sandbox` or `agents.list[].sandbox`). If sandboxing is off, tools run on the host. The Gateway stays on the host; tool execution runs in an isolated sand... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/sandboxing.md | main | opebclaw | [
0.04140090197324753,
0.009512435644865036,
-0.1045290008187294,
0.031766295433044434,
0.08146166056394577,
-0.06321895867586136,
-0.0808921828866005,
-0.02167215757071972,
0.0076998453587293625,
-0.00615054601803422,
0.026560362428426743,
0.022254491224884987,
0.030925871804356575,
-0.0480... | 0.110106 |
ssh: { target: "user@gateway-host:22", workspaceRoot: "/tmp/openclaw-sandboxes", strictHostKeyChecking: true, updateHostKeys: true, identityFile: "~/.ssh/id\_ed25519", certificateFile: "~/.ssh/id\_ed25519-cert.pub", knownHostsFile: "~/.ssh/known\_hosts", // Or use SecretRefs / inline contents instead of local files: //... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/sandboxing.md | main | opebclaw | [
0.0005001731915399432,
0.0838242620229721,
-0.059240106493234634,
0.07238192856311798,
0.056071165949106216,
-0.08996366709470749,
-0.022291945293545723,
-0.04738954082131386,
0.033105626702308655,
0.04419008642435074,
0.013471308164298534,
-0.07973834872245789,
0.05206313356757164,
0.0056... | -0.007001 |
most in practice. ##### `mirror` Use `plugins.entries.openshell.config.mode: "mirror"` when you want the \*\*local workspace to stay canonical\*\*. Behavior: - Before `exec`, OpenClaw syncs the local workspace into the OpenShell sandbox. - After `exec`, OpenClaw syncs the remote workspace back to the local workspace. -... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/sandboxing.md | main | opebclaw | [
0.018435930833220482,
-0.03528665006160736,
-0.001312275999225676,
0.0893728956580162,
0.0471133217215538,
-0.030184922739863396,
-0.11879190802574158,
-0.042539916932582855,
0.022404814139008522,
0.03651757910847664,
0.02584759332239628,
-0.019065624102950096,
0.022252187132835388,
-0.053... | -0.026551 |
be read. With `"rw"`, workspace skills are readable from `/workspace/skills`. ## Custom bind mounts `agents.defaults.sandbox.docker.binds` mounts additional host directories into the container. Format: `host:container:mode` (e.g., `"/home/user/source:/source:rw"`). Global and per-agent binds are \*\*merged\*\* (not rep... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/sandboxing.md | main | opebclaw | [
-0.0195753313601017,
0.0082379300147295,
-0.020982518792152405,
0.059579845517873764,
0.045701753348112106,
-0.03583864867687225,
-0.01651190221309662,
-0.04685850441455841,
-0.028867362067103386,
-0.024624813348054886,
0.018235236406326294,
0.018042264506220818,
0.01940006949007511,
0.060... | 0.026224 |
defaults: - `network: "host"` is blocked. - `network: "container:"` is blocked by default (namespace join bypass risk). - Break-glass override: `agents.defaults.sandbox.docker.dangerouslyAllowContainerNamespaceJoin: true`. Docker installs and the containerized gateway live here: [Docker](/install/docker) For Docker gat... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/sandboxing.md | main | opebclaw | [
0.04274502396583557,
0.011336158029735088,
-0.0364050455391407,
0.0029933408368378878,
0.026116004213690758,
-0.026000486686825752,
-0.05176274850964546,
-0.05414969101548195,
-0.02969682775437832,
-0.019152797758579254,
-0.011128545738756657,
-0.004872299265116453,
0.030136823654174805,
0... | -0.006116 |
# Multiple Gateways (same host) Most setups should use one Gateway because a single Gateway can handle multiple messaging connections and agents. If you need stronger isolation or redundancy (e.g., a rescue bot), run separate Gateways with isolated profiles/ports. ## Isolation checklist (required) - `OPENCLAW\_CONFIG\_... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/multiple-gateways.md | main | opebclaw | [
-0.023770436644554138,
-0.07894302904605865,
-0.10258892178535461,
0.015055692754685879,
-0.012995464727282524,
-0.07753469794988632,
-0.052408333867788315,
0.008912240155041218,
-0.03894542530179024,
-0.04957606643438339,
0.016459424048662186,
-0.016777561977505684,
0.06609027087688446,
0... | 0.042014 |
# Tools Invoke (HTTP) OpenClaw’s Gateway exposes a simple HTTP endpoint for invoking a single tool directly. It is always enabled and uses Gateway auth plus tool policy. Like the OpenAI-compatible `/v1/\*` surface, shared-secret bearer auth is treated as trusted operator access for the whole gateway. - `POST /tools/inv... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/tools-invoke-http-api.md | main | opebclaw | [
-0.08528649806976318,
0.011075573042035103,
-0.116787850856781,
0.03135604411363602,
-0.022781530395150185,
-0.13755927979946136,
-0.020418087020516396,
-0.0008311626152135432,
0.015058540739119053,
0.006065297406166792,
0.007538612000644207,
-0.048580337315797806,
0.030309466645121574,
0.... | 0.055199 |
subagent session key) If a tool is not allowed by policy, the endpoint returns \*\*404\*\*. Important boundary notes: - Exec approvals are operator guardrails, not a separate authorization boundary for this HTTP endpoint. If a tool is reachable here via Gateway auth + tool policy, `/tools/invoke` does not add an extra ... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/tools-invoke-http-api.md | main | opebclaw | [
-0.08212441205978394,
-0.0002697027812246233,
-0.04388287663459778,
-0.04153991863131523,
-0.01333854254335165,
-0.07710213214159012,
-0.0010214647045359015,
0.023279229179024696,
0.06742449849843979,
0.06371194124221802,
0.0004870342672802508,
-0.05087054893374443,
0.021821163594722748,
-... | 0.05319 |
# Heartbeat (Gateway) > \*\*Heartbeat vs Cron?\*\* See [Automation & Tasks](/automation) for guidance on when to use each. Heartbeat runs \*\*periodic agent turns\*\* in the main session so the model can surface anything that needs attention without spamming you. Heartbeat is a scheduled main-session turn — it does \*\... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/heartbeat.md | main | opebclaw | [
-0.0366000197827816,
-0.0007799197919666767,
-0.035810232162475586,
-0.02798372320830822,
0.034178268164396286,
-0.04168441891670227,
-0.012043514288961887,
-0.052205782383680344,
0.02948269434273243,
-0.06267569214105606,
-0.01942276954650879,
-0.082236647605896,
0.04659508913755417,
-0.0... | 0.122547 |
`ackMaxChars`\*\* (default: 300). - If `HEARTBEAT\_OK` appears in the \*\*middle\*\* of a reply, it is not treated specially. - For alerts, \*\*do not\*\* include `HEARTBEAT\_OK`; return only the alert text. Outside heartbeats, stray `HEARTBEAT\_OK` at the start/end of a message is stripped and logged; a message that i... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/heartbeat.md | main | opebclaw | [
-0.08204668760299683,
0.061397530138492584,
0.02449835278093815,
0.059745870530605316,
0.07506733387708664,
-0.07312634587287903,
-0.01339981984347105,
-0.0766199603676796,
0.07495370507240295,
-0.03431106358766556,
0.06680019944906235,
-0.03622761741280556,
-0.0219326913356781,
0.03615172... | 0.119813 |
}, }, }, }, } ``` ### Field notes - `every`: heartbeat interval (duration string; default unit = minutes). - `model`: optional model override for heartbeat runs (`provider/model`). - `includeReasoning`: when enabled, also deliver the separate `Reasoning:` message when available (same shape as `/reasoning on`). - `light... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/heartbeat.md | main | opebclaw | [
-0.08833283931016922,
0.020304203033447266,
-0.07276135683059692,
0.040549036115407944,
0.038642268627882004,
-0.06125515699386597,
0.014852821826934814,
-0.0034901818726211786,
0.01894003339111805,
-0.06385363638401031,
0.04291308671236038,
-0.0007167470175772905,
0.0066391341388225555,
-... | 0.140066 |
run the heartbeat, update due-task timestamps, restore the session idle timestamp, and suppress the outward alert payload. - Heartbeat-only replies do \*\*not\*\* keep the session alive; the last `updatedAt` is restored so idle expiry behaves normally. - Detached [background tasks](/automation/tasks) can enqueue a syst... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/heartbeat.md | main | opebclaw | [
-0.01810535229742527,
0.026807978749275208,
0.01834135875105858,
0.03582487255334854,
0.06976749002933502,
-0.06419757008552551,
0.03658561408519745,
-0.05149071663618088,
0.07982320338487625,
0.01099744625389576,
-0.035706762224435806,
-0.07463234663009644,
-0.00971731636673212,
0.0322361... | 0.082295 |
reply HEARTBEAT\_OK. ``` Behavior: - OpenClaw parses the `tasks:` block and checks each task against its own `interval`. - Only \*\*due\*\* tasks are included in the heartbeat prompt for that tick. - If no tasks are due, the heartbeat is skipped entirely (`reason=no-tasks-due`) to avoid a wasted model call. - Non-task ... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/heartbeat.md | main | opebclaw | [
-0.039568159729242325,
-0.013295541517436504,
-0.02508571743965149,
0.0005506909219548106,
0.04222438856959343,
-0.07514070719480515,
-0.034534793347120285,
-0.06972983479499817,
0.03087189793586731,
-0.0048732697032392025,
-0.0323014073073864,
0.003858993761241436,
0.008871808648109436,
-... | 0.107338 |
# Gateway-owned pairing (Option B) In Gateway-owned pairing, the \*\*Gateway\*\* is the source of truth for which nodes are allowed to join. UIs (macOS app, future clients) are just frontends that approve or reject pending requests. \*\*Important:\*\* WS nodes use \*\*device pairing\*\* (role `node`) during `connect`. ... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/pairing.md | main | opebclaw | [
-0.11740301549434662,
0.025537656620144844,
0.0016892103012651205,
0.025556210428476334,
-0.0031270806211978197,
-0.04167254641652107,
0.06102515757083893,
-0.021129626780748367,
0.047360122203826904,
0.03647146373987198,
-0.026577934622764587,
0.034307487308979034,
0.035446055233478546,
0... | 0.110657 |
dropped, not deferred. ## Node event trust boundaries (2026.3.31+) \*\*Breaking change:\*\* Node-originated runs now stay on a reduced trusted surface. Node-originated summaries and related session events are restricted to the intended trusted surface. Notification-driven or node-triggered flows that previously relied ... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/pairing.md | main | opebclaw | [
-0.06520136445760727,
-0.00645609013736248,
0.050671081990003586,
0.03887090086936951,
0.10278122127056122,
-0.08661002665758133,
0.01425791997462511,
-0.05078306049108505,
0.056870538741350174,
0.05835436284542084,
0.045172590762376785,
0.007005239371210337,
0.02979249320924282,
0.0158966... | 0.084493 |
# OpenResponses API (HTTP) OpenClaw’s Gateway can serve an OpenResponses-compatible `POST /v1/responses` endpoint. This endpoint is \*\*disabled by default\*\*. Enable it in config first. - `POST /v1/responses` - Same port as the Gateway (WS + HTTP multiplex): `http://:/v1/responses` Under the hood, requests are execut... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/openresponses-http-api.md | main | opebclaw | [
-0.04133926331996918,
-0.04065287485718727,
-0.06536931544542313,
0.058858923614025116,
-0.02259923331439495,
-0.09920869022607803,
-0.04105482995510101,
-0.030580410733819008,
0.04645638167858124,
-0.015043356455862522,
-0.01993446983397007,
-0.0011186781339347363,
-0.006902650464326143,
... | 0.055619 |
results back to the model: ```json { "type": "function\_call\_output", "call\_id": "call\_123", "output": "{\"temperature\": \"72F\"}" } ``` ### `reasoning` and `item\_reference` Accepted for schema compatibility but ignored when building the prompt. ## Tools (client-side function tools) Provide tools with `tools: [{ t... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/openresponses-http-api.md | main | opebclaw | [
-0.06288730353116989,
0.003275979310274124,
-0.03191203996539116,
0.0909866914153099,
0.009900910779833794,
-0.03877832368016243,
-0.06426528096199036,
0.05502219870686531,
0.020691053941845894,
-0.010472825728356838,
0.013021509163081646,
-0.036821331828832626,
0.05543629825115204,
0.0694... | 0.046524 |
internet-exposed gateways, apply network egress controls in addition to app-level guards. See [Security](/gateway/security). ## Streaming (SSE) Set `stream: true` to receive Server-Sent Events (SSE): - `Content-Type: text/event-stream` - Each event line is `event: ` and `data: ` - Stream ends with `data: [DONE]` Event ... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/openresponses-http-api.md | main | opebclaw | [
-0.0510382242500782,
0.00851324014365673,
-0.07001891732215881,
0.0909414067864418,
0.03215986490249634,
-0.040762681514024734,
0.09456460922956467,
0.010319139808416367,
0.06963597238063812,
0.024808693677186966,
-0.028309816494584084,
-0.07192289084196091,
-0.03423872962594032,
0.0189298... | 0.14386 |
# Sandbox vs Tool Policy vs Elevated OpenClaw has three related (but different) controls: 1. \*\*Sandbox\*\* (`agents.defaults.sandbox.\*` / `agents.list[].sandbox.\*`) decides \*\*where tools run\*\* (Docker vs host). 2. \*\*Tool policy\*\* (`tools.\*`, `tools.sandbox.tools.\*`, `agents.list[].tools.\*`) decides \*\*w... | https://github.com/openclaw/openclaw/blob/main//docs/gateway/sandbox-vs-tool-policy-vs-elevated.md | main | opebclaw | [
0.007220152765512466,
0.013893699273467064,
-0.04816370829939842,
0.042052753269672394,
0.07546408474445343,
-0.06253159791231155,
-0.005712929181754589,
-0.048871543258428574,
0.015400607138872147,
0.03444225713610649,
0.030363043770194054,
-0.03669708967208862,
0.03980555385351181,
-0.02... | 0.115207 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.