content large_stringlengths 3 20.5k | url large_stringlengths 53 192 ⌀ | branch large_stringclasses 4
values | source large_stringclasses 51
values | embeddings listlengths 384 384 | score float64 -0.21 0.65 |
|---|---|---|---|---|---|
catalog instead of showing an empty provider-scoped picker. - `byteplus/seed-1-8-251228` (Seed 1.8) - `byteplus/kimi-k2-5-260127` (Kimi K2.5) - `byteplus/glm-4-7-251222` (GLM 4.7) - `byteplus-plan/ark-code-latest` - `byteplus-plan/doubao-seed-code` - `byteplus-plan/kimi-k2.5` - `byteplus-plan/kimi-k2-thinking` - `bytep... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/model-providers.md | main | opebclaw | [
-0.06828492879867554,
-0.016096921637654305,
-0.03945939242839813,
-0.01617523841559887,
0.04883476719260216,
-0.0524882897734642,
-0.04959207773208618,
-0.02712518349289894,
-0.013212796300649643,
-0.0009493095567449927,
0.030390625819563866,
-0.07016630470752716,
-0.03566666319966316,
0.... | -0.002713 |
primary: "sglang/your-model-id" } }, }, } ``` See [/providers/sglang](/providers/sglang) for details. ### Local proxies (LM Studio, vLLM, LiteLLM, etc.) Example (OpenAI-compatible): ```json5 { agents: { defaults: { model: { primary: "lmstudio/my-local-model" }, models: { "lmstudio/my-local-model": { alias: "Local" } },... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/model-providers.md | main | opebclaw | [
0.005490352399647236,
-0.039274170994758606,
0.0005163829191587865,
0.023861948400735855,
0.05029105395078659,
0.03423265367746353,
-0.08043801784515381,
-0.02044411189854145,
0.03760415315628052,
-0.008334396407008171,
0.028312167152762413,
-0.05489209294319153,
0.008156265132129192,
0.03... | 0.045225 |
Typing indicators are sent to the chat channel while a run is active. Use `agents.defaults.typingMode` to control \*\*when\*\* typing starts and `typingIntervalSeconds` to control \*\*how often\*\* it refreshes. ## Defaults When `agents.defaults.typingMode` is \*\*unset\*\*, OpenClaw keeps the legacy behavior: - \*\*Di... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/typing-indicators.md | main | opebclaw | [
-0.04041823744773865,
-0.0736704170703888,
0.0011780592612922192,
0.09515389055013657,
0.005614612251520157,
-0.0026326163206249475,
0.08040163666009903,
-0.018357975408434868,
0.07535967975854874,
-0.051120687276124954,
0.022247737273573875,
-0.03855830803513527,
-0.017042528837919235,
0.... | 0.178536 |
OpenClaw handles failures in two stages: 1. \*\*Auth profile rotation\*\* within the current provider. 2. \*\*Model fallback\*\* to the next model in `agents.defaults.model.fallbacks`. This doc explains the runtime rules and the data that backs them. ## Runtime flow For a normal text run, OpenClaw evaluates candidates ... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/model-failover.md | main | opebclaw | [
-0.047674864530563354,
0.0006157025927677751,
-0.02876228466629982,
0.06941530853509903,
0.023242462426424026,
-0.05834019184112549,
-0.02488887496292591,
0.022580387070775032,
-0.035889338701963425,
-0.021374646574258804,
0.01403826754540205,
-0.0237734355032444,
0.07120630145072937,
0.01... | -0.031974 |
the selected primary; it does not repeat the notice on every sticky fallback turn. ## Auth failure skip cache By default, every new turn keeps the existing fallback retry behavior: OpenClaw will try each configured fallback candidate again, including non-primary candidates that recently failed with `auth` or `auth\_per... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/model-failover.md | main | opebclaw | [
-0.046519383788108826,
0.015063601545989513,
-0.052773699164390564,
0.09064320474863052,
0.059753648936748505,
-0.07088575512170792,
0.026786256581544876,
-0.01844274066388607,
0.03807314112782478,
-0.07166512310504913,
0.03054475039243698,
0.0011853621108457446,
0.04972858726978302,
-0.03... | 0.031161 |
without changing the selected model or runtime. User-pinned profiles stay locked to that profile; if it fails and model fallbacks are configured, OpenClaw moves to the next model instead of switching profiles. ### OpenAI Codex subscription plus API-key backup For OpenAI agent models, auth and runtime are separate. `ope... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/model-failover.md | main | opebclaw | [
-0.07303030043840408,
-0.0370844230055809,
-0.04865797609090805,
0.04840359836816788,
0.044662561267614365,
-0.03540021926164627,
0.010738825425505638,
-0.0008137337863445282,
0.019896134734153748,
-0.051191527396440506,
0.026950232684612274,
0.035136234015226364,
0.042572151869535446,
0.0... | -0.01032 |
profile across models. Cooldowns use exponential backoff: - 1 minute - 5 minutes - 25 minutes - 1 hour (cap) State is stored in `auth-state.json` under `usageStats`: ```json { "usageStats": { "provider:profile": { "lastUsed": 1736160000000, "cooldownUntil": 1736160600000, "errorCount": 2 } } } ``` ## Billing disables B... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/model-failover.md | main | opebclaw | [
-0.09636921435594559,
0.051030825823545456,
-0.0446610227227211,
0.09805922210216522,
0.006033803801983595,
-0.07192804664373398,
-0.0155453747138381,
0.026082005351781845,
0.04755845293402672,
-0.04750848934054375,
0.032579950988292694,
-0.002408594824373722,
0.03494420647621155,
0.011799... | -0.009191 |
requested model uses a different provider. - When no explicit fallback override is supplied to the fallback runner, the configured primary is appended at the end so the chain can settle back onto the normal default once earlier candidates are exhausted. - When a caller supplies `fallbacksOverride`, the runner uses exac... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/model-failover.md | main | opebclaw | [
-0.05536068603396416,
-0.02881506457924843,
0.03643494471907616,
0.07742099463939667,
0.017426781356334686,
-0.028874089941382408,
-0.06665044277906418,
0.05495661497116089,
-0.02421782910823822,
0.005939427297562361,
0.0328885018825531,
0.03202495351433754,
0.0530109740793705,
-0.05412423... | 0.034545 |
If a live-switch error points at a later candidate in the active fallback chain, OpenClaw jumps directly to that selected model instead of walking unrelated candidates first. - If the fallback attempt fails, the runner rolls back only the override fields it wrote, and only if they still match that failed candidate. Thi... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/model-failover.md | main | opebclaw | [
-0.06254277378320694,
0.003507677000015974,
-0.010974983684718609,
0.08800160884857178,
0.0669122114777565,
-0.03630680590867996,
-0.0731479823589325,
0.030674120411276817,
-0.02166171185672283,
0.02629810944199562,
0.03652200475335121,
0.03198981285095215,
0.006355226505547762,
-0.0337169... | -0.000802 |
OpenClaw remembers things by writing \*\*plain Markdown files\*\* in your agent's workspace. The model only "remembers" what gets saved to disk — there is no hidden state. ## How it works Your agent has three memory-related files: - \*\*`MEMORY.md`\*\* — long-term memory. Durable facts, preferences, and decisions. Load... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/memory.md | main | opebclaw | [
-0.04083046689629555,
-0.0026177740655839443,
-0.06906724721193314,
0.10183485597372055,
0.019254688173532486,
-0.002717014867812395,
-0.04221062362194061,
-0.029850279912352562,
0.04463469982147217,
-0.010542506352066994,
-0.0013677674578502774,
0.09116146713495255,
-0.018018081784248352,
... | 0.092923 |
untrusted source needs review before promotion. Future turns should treat it as evidence only; do not store it as durable memory until a trusted reviewer confirms the contents. ``` Use [commitments](/concepts/commitments) for inferred, short-lived follow-ups. Use [scheduled tasks](/automation/cron-jobs) for exact remin... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/memory.md | main | opebclaw | [
-0.10888568311929703,
0.07650476694107056,
-0.1493905484676361,
0.04981658235192299,
0.08716560155153275,
-0.0066922749392688274,
0.02370467223227024,
-0.018275100737810135,
0.03463498875498772,
0.0013077984331175685,
0.008150496520102024,
0.02571442909538746,
0.03854827210307121,
0.015622... | 0.108862 |
does not inherit the active session fallback chain. The memory flush prevents context loss during compaction. If your agent has important facts in the conversation that are not yet written to a file, they will be saved automatically before the summary happens. ## Dreaming Dreaming is an optional background consolidatio... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/memory.md | main | opebclaw | [
0.010021531023085117,
-0.003085761098191142,
-0.0807848647236824,
0.07578718662261963,
0.03328903391957283,
0.04694065824151039,
0.011998595669865608,
-0.023843973875045776,
0.03611823916435242,
-0.012157024815678596,
-0.02770250476896763,
0.08602236956357956,
0.014422341249883175,
-0.0338... | 0.091372 |
Progress drafts make long-running agent turns feel alive in chat without turning the conversation into a stack of temporary status replies. When progress drafts are enabled, OpenClaw creates one visible work-in-progress message only after the turn proves it is doing real work, updates it while the agent reads, plans, c... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/progress-drafts.md | main | opebclaw | [
-0.034702662378549576,
-0.013071617111563683,
-0.015693042427301407,
0.10032688826322556,
0.028043633326888084,
-0.06667768955230713,
-0.03142121434211731,
-0.06075720861554146,
0.08422576636075974,
-0.0001904726232169196,
-0.037211477756500244,
0.028577078133821487,
-0.04439983144402504,
... | 0.1012 |
Nautiling Krilling Barnacling Lobstering Tidepooling Pearling Snapping Surfacing ``` Use a fixed label: ```json5 { channels: { discord: { streaming: { mode: "progress", progress: { label: "Investigating", }, }, }, }, } ``` Use your own automatic label pool: ```json5 { channels: { discord: { streaming: { mode: "progress... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/progress-drafts.md | main | opebclaw | [
-0.05273013189435005,
0.05493060499429703,
0.04966219887137413,
0.006242497358471155,
-0.013032712042331696,
0.03713731840252876,
0.01368613913655281,
-0.06405480206012726,
-0.011397380381822586,
-0.03536562994122505,
-0.01352434791624546,
-0.0736132264137268,
-0.04211180657148361,
0.07413... | 0.13066 |
}, }, }, }, } ``` Slack can render progress lines as structured Block Kit fields instead of a single text body: ```json5 { channels: { slack: { streaming: { mode: "progress", progress: { render: "rich", }, }, }, }, } ``` Rich rendering keeps the same plain-text fallback so channels and clients that do not support the r... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/progress-drafts.md | main | opebclaw | [
-0.07106691598892212,
0.06650122255086899,
-0.0017777075991034508,
0.06830772757530212,
0.03834151849150658,
-0.034292835742235184,
-0.06429826468229294,
-0.06121661886572838,
-0.0007143319235183299,
-0.033664271235466,
-0.021731307730078697,
-0.007155234459787607,
-0.033317361027002335,
0... | 0.022728 |
`streaming.mode: "off"` or a channel path that cannot create a draft for that message. \*\*Teams behaves differently from Discord or Telegram.\*\* Microsoft Teams uses a native stream in personal chats instead of the generic send-and-edit preview transport. Teams also treats `streaming.mode: "block"` as Teams block del... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/progress-drafts.md | main | opebclaw | [
-0.022280652076005936,
-0.03350228816270828,
-0.01287044957280159,
0.013658509589731693,
0.053321946412324905,
-0.035968177020549774,
-0.0010410072281956673,
-0.0787869542837143,
0.06421681493520737,
0.03340524807572365,
-0.012373631820082664,
-0.0021752251777797937,
-0.025906966999173164,
... | 0.092747 |
TypeBox is a TypeScript-first schema library. We use it to define the \*\*Gateway WebSocket protocol\*\* (handshake, request/response, server events). Those schemas drive \*\*runtime validation\*\*, \*\*JSON Schema export\*\*, and \*\*Swift codegen\*\* for the macOS app. One source of truth; everything else is generate... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/typebox.md | main | opebclaw | [
0.006331244483590126,
-0.033774636685848236,
0.003609700361266732,
0.0392456129193306,
-0.020285500213503838,
-0.02325470931828022,
0.018336107954382896,
-0.015253642573952675,
0.011757437139749527,
-0.0573868453502655,
-0.09745663404464722,
-0.02533181570470333,
-0.07692108303308487,
0.06... | 0.16422 |
"event", "event": "tick", "payload": { "ts": 1730000000 }, "seq": 12 } ``` ## Minimal client (Node.js) Smallest useful flow: connect + health. ```ts import { WebSocket } from "ws"; const ws = new WebSocket("ws://127.0.0.1:18789"); ws.on("open", () => { ws.send( JSON.stringify({ type: "req", id: "c1", method: "connect",... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/typebox.md | main | opebclaw | [
-0.05712670460343361,
0.05076489970088005,
0.009234069846570492,
0.046516090631484985,
-0.037433456629514694,
-0.0433834008872509,
0.07205699384212494,
0.0923846960067749,
0.0682452917098999,
-0.06343713402748108,
-0.046186063438653946,
0.002493060426786542,
-0.0032119518145918846,
0.00941... | 0.120217 |
Auth profile rotation, cooldowns, and how that interacts with fallbacks. Quick provider overview and examples. OpenClaw, Codex, and other agent loop runtimes. Model config keys. Model refs choose a provider and model. They do not usually choose the low-level agent runtime. OpenAI agent refs are the main exception: `ope... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/models.md | main | opebclaw | [
-0.11015249043703079,
-0.06295568495988846,
-0.11640441417694092,
0.028225449845194817,
0.03435878828167915,
-0.0491359569132328,
0.016760218888521194,
0.07970495522022247,
0.03695995733141899,
-0.03456566482782364,
0.05144362896680832,
0.029820242896676064,
0.03933534026145935,
0.02005544... | 0.066174 |
payload `fallbacks` (use `fallbacks: []` for a strict cron run). - CLI default-model and allowlist pickers respect `models.mode: "replace"` by listing explicit `models.providers.\*.models` instead of loading the full built-in catalog. - The Control UI model picker asks the Gateway for its configured model view: `agents... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/models.md | main | opebclaw | [
-0.07548685371875763,
-0.09674467891454697,
-0.06278781592845917,
0.08717943727970123,
0.009219889529049397,
0.0026926109567284584,
-0.04118940979242325,
0.03458191826939583,
-0.04271344095468521,
-0.03439609333872795,
0.030456235632300377,
-0.0765090063214302,
0.004337376449257135,
0.0552... | 0.069352 |
manually listing every model, add `provider/\*` entries to `agents.defaults.models`: ```json5 { agents: { defaults: { models: { "openai/\*": {}, "vllm/\*": {}, }, }, }, } ``` With that policy, `/model`, `/models`, and model pickers show the discovered catalog for those providers only. New models from the selected provi... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/models.md | main | opebclaw | [
-0.0389384888112545,
-0.039939045906066895,
0.04274366796016693,
0.0489206500351429,
0.04041368514299393,
0.017191998660564423,
-0.009879732504487038,
-0.048352230340242386,
-0.02224241942167282,
0.0006223978707566857,
0.032052841037511826,
-0.0679459199309349,
-0.06468486785888672,
0.1344... | 0.029228 |
Local providers only. Filter by provider id, for example `moonshot`. Display labels from interactive pickers are not accepted. One model per line. Machine-readable output. ### `models status` Shows the resolved primary model, fallbacks, image model, and an auth overview of configured providers. It also surfaces OAuth e... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/models.md | main | opebclaw | [
-0.02806086838245392,
0.05049707740545273,
-0.03689063340425491,
0.054408296942710876,
0.1106177493929863,
-0.030950741842389107,
0.015142153948545456,
-0.013093283399939537,
-0.016964087262749672,
-0.0620243102312088,
0.07220087200403214,
-0.045495375990867615,
0.05267171189188957,
0.0305... | -0.044962 |
fall back to config `models.providers`. - Other provider fields are refreshed from config and normalized catalog data. Marker persistence is source-authoritative: OpenClaw writes markers from the active source config snapshot (pre-resolution), not from resolved runtime secret values. This applies whenever OpenClaw rege... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/models.md | main | opebclaw | [
-0.01836409606039524,
-0.0756097361445427,
-0.07278579473495483,
0.08679158240556717,
0.0778401792049408,
-0.06348711997270584,
-0.10409723222255707,
-0.026188042014837265,
-0.005089045502245426,
-0.0847650021314621,
0.07228408008813858,
0.011577822268009186,
0.006897305604070425,
-0.01342... | 0.019807 |
An agentic loop is the full "real" run of an agent: intake → context assembly → model inference → tool execution → streaming replies → persistence. It's the authoritative path that turns a message into actions and a final reply, while keeping session state consistent. In OpenClaw, a loop is a single, serialized run per... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/agent-loop.md | main | opebclaw | [
-0.018294408917427063,
-0.04607642814517021,
-0.16448095440864563,
0.01218650583177805,
0.029192714020609856,
-0.08468246459960938,
0.018719924613833427,
-0.00651067728176713,
0.045782655477523804,
-0.03304503485560417,
0.026096567511558533,
0.006352362688630819,
-0.005090692546218634,
-0.... | 0.120463 |
pipeline. ### Internal hooks (Gateway hooks) - \*\*`agent:bootstrap`\*\*: runs while building bootstrap files before the system prompt is finalized. Use this to add/remove bootstrap context files. - \*\*Command hooks\*\*: `/new`, `/reset`, `/stop`, and other command events (see Hooks doc). See [Hooks](/automation/hooks... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/agent-loop.md | main | opebclaw | [
-0.025934861972928047,
-0.005209608469158411,
-0.03432188928127289,
0.04036413133144379,
-0.008854835294187069,
-0.007188170216977596,
-0.02343147061765194,
0.05939009413123131,
-0.018913786858320236,
0.031597867608070374,
-0.0009973666165024042,
-0.023528626188635826,
0.023950830101966858,
... | 0.058759 |
`compaction` stream events and can trigger a retry. - On retry, in-memory buffers and tool summaries are reset to avoid duplicate output. - See [Compaction](/concepts/compaction) for the compaction pipeline. ## Event streams (today) - `lifecycle`: emitted by `subscribeEmbeddedAgentSession` (and as a fallback by `agentC... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/agent-loop.md | main | opebclaw | [
-0.026526255533099174,
-0.02964070253074169,
-0.0664965882897377,
0.009571345522999763,
-0.03904832899570465,
-0.023541845381259918,
0.012083771638572216,
-0.012423803098499775,
0.04403098672628403,
-0.0030228656250983477,
0.025580447167158127,
-0.019487185403704643,
-0.03475693613290787,
... | 0.173095 |
OpenClaw formats outbound Markdown by converting it into a shared intermediate representation (IR) before rendering channel-specific output. The IR keeps the source text intact while carrying style/link spans so chunking and rendering can stay consistent across channels. ## Goals - \*\*Consistency:\*\* one parse step, ... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/markdown-formatting.md | main | opebclaw | [
-0.0412774384021759,
0.033057063817977905,
0.028327632695436478,
0.029345091432332993,
0.054567065089941025,
-0.07388999313116074,
-0.07440265268087387,
-0.0016519749769940972,
0.015093985944986343,
0.003012818517163396,
-0.07899630814790726,
-0.020131077617406845,
-0.013078340329229832,
-... | -0.001861 |
use the new chunker and renderer. 5. **Test:** add or update format tests and an outbound delivery test if the channel uses chunking. ## Common gotchas - Slack angle-bracket tokens (`<@U123>`, `<#C123>`, ``) must be preserved; escape raw HTML safely. - Telegram HTML requires escaping text outside tags to avoid broken m... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/markdown-formatting.md | main | opebclaw | [
-0.06394273787736893,
-0.012737014330923557,
0.04719102382659912,
-0.01907288283109665,
0.017809392884373665,
-0.06693269312381744,
-0.040336836129426956,
-0.016661955043673515,
-0.0042226132936775684,
0.021344952285289764,
-0.021776093170046806,
-0.017130156978964806,
-0.031733643263578415,... | 0.109854 |
OpenClaw gives agents tools to work across sessions, inspect status, and orchestrate sub-agents. ## Available tools | Tool | What it does | | ------------------ | --------------------------------------------------------------------------- | | `sessions\_list` | List sessions with optional filters (kind, label, agent, r... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/session-tool.md | main | opebclaw | [
-0.046494774520397186,
-0.017790138721466064,
-0.08146834373474121,
0.04628052935004234,
0.02562795765697956,
-0.06549397855997086,
0.02739657834172249,
-0.02429584600031376,
-0.01347615197300911,
0.012925335206091404,
0.005333119537681341,
-0.056829147040843964,
-0.0034681642428040504,
-0... | 0.116341 |
and return immediately. - \*\*Wait for reply:\*\* set a timeout and get the response inline. Thread-scoped chat sessions, such as Slack or Discord keys ending in `:thread:`, are not valid `sessions\_send` targets. Use the parent channel session key for inter-agent coordination so tool-routed messages do not appear insi... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/session-tool.md | main | opebclaw | [
-0.057498421519994736,
-0.010101737454533577,
-0.05004057660698891,
0.05546296760439873,
-0.01140995230525732,
-0.06445945054292679,
0.039869192987680435,
-0.04979174584150314,
0.02516421116888523,
0.024544138461351395,
0.007658324670046568,
-0.048417408019304276,
-0.020649630576372147,
-0... | 0.097481 |
can still reuse the requester session's stored route (`lastChannel` / `lastTo`) for direct delivery. For ACP-specific behavior, see [ACP Agents](/tools/acp-agents). ## Visibility Session tools are scoped to limit what the agent can see: | Level | Scope | | ------- | ---------------------------------------- | | `self` |... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/session-tool.md | main | opebclaw | [
-0.03262244537472725,
-0.011760040186345577,
-0.05419376492500305,
0.027311574667692184,
0.013912658207118511,
-0.03556741401553154,
0.04445260390639305,
-0.0037989255506545305,
-0.004198359791189432,
0.04049475118517876,
0.004352418705821037,
0.011710419319570065,
0.008852701634168625,
0.... | 0.097823 |
The builtin engine is the default memory backend. It stores your memory index in a per-agent SQLite database and needs no extra dependencies to get started. ## What it provides - \*\*Keyword search\*\* via FTS5 full-text indexing (BM25 scoring). - \*\*Vector search\*\* via embeddings from any supported provider. - \*\*... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/memory-builtin.md | main | opebclaw | [
-0.021033981814980507,
-0.014424164779484272,
-0.08233454823493958,
0.10638003051280975,
0.04034474119544029,
0.011249908246099949,
-0.01472737081348896,
-0.013002678751945496,
-0.01713073067367077,
-0.03263934329152107,
0.02365979552268982,
0.019148027524352074,
0.007884912192821503,
-0.0... | 0.151861 |
automatically. `openclaw memory status --deep` reports the local vector store separately from the embedding provider, so `Vector store: unavailable` points at sqlite-vec loading while `Embeddings: unavailable` points at provider/auth or model readiness. Check logs for the specific load error. ## Configuration For embed... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/memory-builtin.md | main | opebclaw | [
0.026248879730701447,
-0.10910071432590485,
-0.10141133517026901,
0.06954581290483475,
0.01890529692173004,
-0.02100229077041149,
-0.07575006783008575,
0.002228793455287814,
-0.03916241228580475,
-0.013142654672265053,
0.00776998745277524,
0.01887470670044422,
-0.022616596892476082,
-0.019... | 0.033575 |
We serialize inbound auto-reply runs (all channels) through a tiny in-process queue to prevent multiple agent runs from colliding, while still allowing safe parallelism across sessions. ## Why - Auto-reply runs can be expensive (LLM calls) and can collide when multiple inbound messages arrive close together. - Serializ... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/queue.md | main | opebclaw | [
-0.04891445115208626,
-0.04445498436689377,
-0.062483787536621094,
-0.04101936146616936,
-0.07931102812290192,
-0.05097757652401924,
0.03928351774811745,
0.0003274953051004559,
0.04764844849705696,
0.06151397153735161,
-0.02985728159546852,
0.0025238937232643366,
-0.011970575898885727,
-0.... | 0.151887 |
Steer and streaming When channel streaming is `partial` or `block`, steering can look like several short visible replies while the active run reaches runtime boundaries: - `partial`: the preview may finalize early, then a new preview starts after steering is accepted. - `block`: draft-sized blocks can create the same s... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/queue.md | main | opebclaw | [
0.04356899484992027,
-0.07835663110017776,
-0.023084575310349464,
0.041880618780851364,
0.027773944661021233,
-0.08924051374197006,
0.004607738461345434,
-0.03743813559412956,
0.01942051574587822,
-0.01041332259774208,
0.03971722722053528,
0.03109634667634964,
-0.026547404006123543,
-0.031... | -0.017033 |
Channel docking is call forwarding for one OpenClaw session. It keeps the same conversation context, but changes where future replies for that session are delivered. ## Example Alice can message OpenClaw on Telegram and Discord: ```json5 { session: { identityLinks: { alice: ["telegram:123", "discord:456"], }, }, } ``` ... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/channel-docking.md | main | opebclaw | [
-0.06825783848762512,
-0.002819135319441557,
0.01623520255088806,
0.0178794264793396,
0.0193089060485363,
-0.09204704314470291,
0.0944414809346199,
-0.03272123262286186,
0.16548211872577667,
0.022847317159175873,
-0.03740664944052696,
0.024518059566617012,
-0.01187918335199356,
0.052915886... | 0.139015 |
peer id matches the id used by that channel. Docking only changes the active session route; another session may still route elsewhere. \*\*I need to switch back.\*\* Send the matching command for the original channel, such as `/dock\_telegram` or `/dock-telegram`, from a linked sender. | https://github.com/openclaw/openclaw/blob/main/docs/concepts/channel-docking.md | main | opebclaw | [
-0.009082197211682796,
0.03509917110204697,
0.04172735661268234,
-0.033428654074668884,
-0.023853763937950134,
-0.03350464254617691,
0.08785419911146164,
-0.04842744022607803,
0.01449059322476387,
0.05939174443483353,
-0.059568095952272415,
0.015174427069723606,
0.026974912732839584,
0.046... | 0.017168 |
## Highlights Discord, iMessage, Signal, Slack, Telegram, WhatsApp, WebChat, and more with a single Gateway. Bundled plugins add Matrix, Nextcloud Talk, Nostr, Twitch, Zalo, and more without separate installs in normal current releases. Multi-agent routing with isolated sessions. Images, audio, video, documents, and im... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/features.md | main | opebclaw | [
-0.04574170336127281,
-0.12185855209827423,
-0.0016848952509462833,
-0.057284627109766006,
0.01761765033006668,
-0.01422363892197609,
0.048644475638866425,
-0.0008643046021461487,
0.016927329823374748,
0.05232769250869751,
0.05835600197315216,
-0.021581115201115608,
-0.006848481018096209,
... | 0.187684 |
`SOUL.md` is where your agent's voice lives. OpenClaw injects it on normal sessions, so it has real weight. If your agent sounds bland, hedgy, or weirdly corporate, this is usually the file to fix. ## What belongs in SOUL.md Put the stuff that changes how the agent feels to talk to: - tone - opinions - brevity - humor ... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/soul.md | main | opebclaw | [
0.019870491698384285,
0.0166695024818182,
-0.028674306347966194,
-0.02101333625614643,
0.027385229244828224,
-0.06885351985692978,
0.03156411275267601,
-0.01491916086524725,
0.0221878495067358,
-0.06029783561825752,
-0.031630739569664,
0.007310282438993454,
-0.016312463209033012,
-0.064266... | 0.105265 |
Codex runtime context. Starter template for a personality file. | https://github.com/openclaw/openclaw/blob/main/docs/concepts/soul.md | main | opebclaw | [
-0.026122653856873512,
0.08058531582355499,
-0.07301564514636993,
0.02750173583626747,
0.03697545453906059,
-0.006645532790571451,
0.07246240228414536,
0.11364040523767471,
-0.1303226798772812,
-0.03724909573793411,
0.007135006599128246,
-0.05561462417244911,
0.006221161223948002,
-0.04114... | 0.154234 |
OpenClaw organizes conversations into \*\*sessions\*\*. Each message is routed to a session based on where it came from -- DMs, group chats, cron jobs, etc. ## How messages are routed | Source | Behavior | | --------------- | ------------------------- | | Direct messages | Shared session by default | | Group chats | Is... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/session.md | main | opebclaw | [
-0.05818106234073639,
-0.016180863603949547,
-0.05151380971074104,
0.049886252731084824,
0.014806408435106277,
-0.13790446519851685,
0.045076481997966766,
-0.06675146520137787,
0.027843043208122253,
0.008443397469818592,
0.028590619564056396,
-0.022104483097791672,
0.0026881699450314045,
0... | 0.0685 |
during maintenance. Set `session.maintenance.mode` to `"warn"` to report what would be cleaned without mutating the store/files: ```json5 { session: { maintenance: { mode: "enforce", pruneAfter: "30d", maxEntries: 500, }, }, } ``` For production-sized `maxEntries` limits, Gateway runtime writes use a small high-water b... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/session.md | main | opebclaw | [
-0.043875016272068024,
0.05288049206137657,
0.0034809280186891556,
0.033226147294044495,
0.02802588790655136,
-0.11501553654670715,
0.02288665995001793,
-0.027222488075494766,
-0.025724543258547783,
0.06270056217908859,
-0.0017921016551554203,
0.10090920329093933,
-0.05498562380671501,
0.0... | 0.06023 |
Goal: run OpenClaw as a \*\*named delegate\*\* - an agent with its own identity that acts "on behalf of" people in an organization. The agent never impersonates a human. It sends, reads, and schedules under its own account with explicit delegation permissions. This extends [Multi-Agent Routing](/concepts/multi-agent) f... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/delegate-architecture.md | main | opebclaw | [
-0.06751907616853714,
-0.006512392312288284,
-0.07136920839548111,
-0.00026830617571249604,
-0.024324890226125717,
-0.0716395303606987,
0.02525077946484089,
-0.06845064461231232,
-0.009648611769080162,
0.0230685006827116,
0.026733241975307465,
-0.02451152727007866,
0.01775653101503849,
0.0... | 0.135872 |
the agent \*\*cannot\*\* do. Establish these constraints before giving it the ability to do anything. ### Hard blocks (non-negotiable) Define these in the delegate's `SOUL.md` and `AGENTS.md` before connecting any external accounts: - Never send external emails without explicit human approval. - Never export contact li... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/delegate-architecture.md | main | opebclaw | [
-0.05370229855179787,
0.018309706822037697,
-0.07693227380514145,
0.02335525117814541,
-0.001993741374462843,
-0.012355973944067955,
-0.0076391007751226425,
-0.07447904348373413,
-0.0012923333561047912,
0.01996142975986004,
0.030526360496878624,
0.022431250661611557,
0.021899281069636345,
... | 0.07314 |
model. Domain-wide delegation allows the service account to impersonate \*\*any user in the entire domain\*\*. Restrict the scopes to the minimum required, and limit the service account's client ID to only the scopes listed above in the Admin Console (Security > API controls > Domain-wide delegation). A leaked service ... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/delegate-architecture.md | main | opebclaw | [
-0.07784636318683624,
-0.013257849961519241,
0.03587054833769798,
0.02912900596857071,
0.016109539195895195,
-0.11155238747596741,
0.045251499861478806,
-0.09078653901815414,
0.08313187956809998,
0.01362334843724966,
-0.04265088215470314,
0.00571462232619524,
-0.010112241841852665,
0.11766... | -0.000571 |
[Honcho](https://honcho.dev) adds AI-native memory to OpenClaw. It persists conversations to a dedicated service and builds user and agent models over time, giving your agent cross-session context that goes beyond workspace Markdown files. ## What it provides - \*\*Cross-session memory\*\* -- conversations are persiste... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/memory-honcho.md | main | opebclaw | [
-0.01029778178781271,
-0.058935634791851044,
-0.04979754239320755,
0.032961420714855194,
0.02730580046772957,
-0.04374644160270691,
0.05124112218618393,
0.009672518819570541,
0.0207176823168993,
-0.008383066393435001,
0.022540278732776642,
-0.03807354345917702,
0.002634286880493164,
-0.016... | 0.276986 |
honcho status # Check connection status openclaw honcho ask # Query Honcho about the user openclaw honcho search [-k N] [-d D] # Semantic search over memory ``` ## Further reading - [Plugin source code](https://github.com/plastic-labs/openclaw-honcho) - [Honcho documentation](https://docs.honcho.dev) - [Honcho OpenClaw... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/memory-honcho.md | main | opebclaw | [
-0.019884511828422546,
-0.013096033595502377,
-0.09898920357227325,
0.040294066071510315,
-0.051733698695898056,
-0.059134677052497864,
-0.03436826542019844,
-0.0026451984886080027,
-0.04602791741490364,
-0.010230278596282005,
0.07615938782691956,
-0.012319321744143963,
0.023952234536409378,... | 0.170749 |
OpenClaw builds a custom system prompt for every agent run. The prompt is \*\*OpenClaw-owned\*\* and does not use a runtime default prompt. The prompt is assembled by OpenClaw and injected into each agent run. Prompt assembly has three layers: - `buildAgentSystemPrompt` renders the prompt from explicit inputs. It shoul... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/system-prompt.md | main | opebclaw | [
0.002367717446759343,
-0.0003622034564614296,
-0.060922615230083466,
0.05260375887155533,
0.06228208169341087,
-0.0877934917807579,
-0.05891139432787895,
0.01038642879575491,
-0.01825794205069542,
-0.017419949173927307,
-0.010995649732649326,
-0.09767555445432663,
0.04885711148381233,
-0.0... | 0.038007 |
accepted schema already carries that runtime detail. The Tooling section also includes runtime guidance for long-running work: - use cron for future follow-up (`check back later`, reminders, recurring work) instead of `exec` sleep loops, `yieldMs` delay tricks, or repeated `process` polling - use `exec` / `process` onl... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/system-prompt.md | main | opebclaw | [
-0.03371705859899521,
-0.03925764188170433,
-0.06532152742147446,
0.03645304962992668,
0.005732598714530468,
-0.0534561425447464,
-0.02196878381073475,
0.012388773262500763,
-0.04680614545941353,
0.02758885733783245,
-0.04078183323144913,
-0.051670756191015244,
-0.045536886900663376,
0.065... | 0.126882 |
sources exist, the command exits without changing the committed fixture. Pass `--catalog ` to refresh from a specific `models\_cache.json` or `models.json` file. These snapshots are still not a byte-for-byte raw OpenAI request capture. Codex can add runtime-owned workspace context such as `AGENTS.md`, environment conte... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/system-prompt.md | main | opebclaw | [
-0.03941497206687927,
-0.0212369617074728,
-0.09131467342376709,
0.03363567218184471,
0.12770293653011322,
-0.09843626618385315,
-0.06030789762735367,
-0.02840348705649376,
0.026996301487088203,
0.019010763615369797,
0.03939126059412956,
-0.029249824583530426,
-0.021370043978095055,
-0.009... | 0.061783 |
model only sees the shortened injected copy until it reads or searches memory directly. If `MEMORY.md` is repeatedly truncated there, distill it into a shorter durable summary and move detailed history into `memory/\*.md`, or intentionally raise the bootstrap limits. Sub-agent sessions only inject `AGENTS.md` and `TOOL... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/system-prompt.md | main | opebclaw | [
0.029697280377149582,
0.03114616870880127,
-0.056898392736911774,
0.007740626577287912,
0.024515904486179352,
-0.018396899104118347,
0.007739298045635223,
0.03060263767838478,
-0.01135617308318615,
-0.005761954467743635,
-0.013379440642893314,
0.05018126592040062,
0.04172438755631447,
-0.0... | 0.083593 |
commands, configuration, or architecture, and to run `openclaw status` itself when possible (asking the user only when it lacks access). For configuration specifically, it points agents to the `gateway` tool action `config.schema.lookup` for exact field-level docs and constraints, then to `docs/gateway/configuration.md... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/system-prompt.md | main | opebclaw | [
0.018760118633508682,
-0.017088625580072403,
-0.14251945912837982,
0.034520626068115234,
-0.005169246811419725,
-0.09269651025533676,
-0.03632400929927826,
-0.002965605352073908,
-0.07395599782466888,
-0.029686547815799713,
0.013078452087938786,
0.0064809322357177734,
0.004032937344163656,
... | 0.102966 |
Experimental features in OpenClaw are \*\*opt-in preview surfaces\*\*. They are behind explicit flags because they still need real-world mileage before they deserve a stable default or a long-lived public contract. Treat them differently from normal config: - Keep them \*\*off by default\*\* unless the related doc tell... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/experimental-features.md | main | opebclaw | [
-0.005423780530691147,
0.04502459615468979,
0.005197819322347641,
0.059887152165174484,
0.03951949626207352,
-0.1173129752278328,
-0.05509745702147484,
-0.043808579444885254,
-0.0602545402944088,
-0.012848915532231331,
0.013652399182319641,
-0.051108263432979584,
-0.021341146901249886,
-0.... | -0.028254 |
Enable ```json5 { agents: { defaults: { experimental: { localModelLean: true, }, }, }, } ``` For one agent only: ```json5 { agents: { list: [ { id: "local", model: "lmstudio/gemma-4-e4b-it", experimental: { localModelLean: true, }, }, ], }, } ``` Restart the Gateway after changing the flag, then confirm the trimmed too... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/experimental-features.md | main | opebclaw | [
-0.0028121822979301214,
-0.009974436834454536,
-0.026231352239847183,
0.07253395766019821,
0.10278977453708649,
-0.11582721024751663,
-0.10328159481287003,
-0.05505651608109474,
-0.08361145853996277,
-0.0170481838285923,
0.07920674979686737,
-0.035840801894664764,
0.0053911334834992886,
-0... | 0.055443 |
Mantis is the OpenClaw end-to-end verification system for bugs that need a real runtime, a real transport, and visible proof. It runs a scenario against a known bad ref, captures evidence, runs the same scenario against a candidate ref, and publishes the comparison as artifacts that a maintainer can inspect from a PR o... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/mantis.md | main | opebclaw | [
-0.06036483868956566,
-0.030433734878897667,
-0.07807765156030655,
-0.021576330065727234,
0.053807247430086136,
-0.07194384932518005,
-0.0012204224476590753,
-0.011354297399520874,
-0.047488871961832047,
0.0491388775408268,
0.036195144057273865,
-0.06895465403795242,
0.016186362132430077,
... | 0.172134 |
\ --candidate \ --output-dir .artifacts/qa-e2e/mantis/local-discord-thread-attachment ``` That scenario posts a parent message with the driver bot, creates a real Discord thread, calls OpenClaw's `message.thread-reply` action with a repo-local `filePath`, then polls the thread for the SUT reply and attachment filename.... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/mantis.md | main | opebclaw | [
-0.016163349151611328,
0.034853387624025345,
-0.08283732086420059,
-0.012930317781865597,
-0.00128409662283957,
-0.07074804604053497,
-0.03100399300456047,
-0.029494255781173706,
0.014098245650529861,
-0.004718957934528589,
0.0756693035364151,
-0.016408409923315048,
-0.036961980164051056,
... | 0.018855 |
and the VNC screenshot back to the local output directory. This is the first Mantis shape where the SUT OpenClaw gateway and the browser both live inside the same Linux desktop VM. With `--gateway-setup`, the command prepares a persistent disposable OpenClaw home at `$HOME/.openclaw-mantis/slack-openclaw`, patches Slac... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/mantis.md | main | opebclaw | [
0.00315615301951766,
-0.04983169212937355,
-0.07413241267204285,
0.015312333591282368,
0.0482756569981575,
-0.04811996594071388,
-0.10671589523553848,
-0.003915904555469751,
-0.05802887678146362,
-0.06564021110534668,
0.05293150618672371,
-0.04352063685655594,
-0.013700810261070728,
0.0228... | -0.04865 |
the generic manual entrypoint. It takes a `scenario\_id`, `candidate\_ref`, optional `baseline\_ref`, and optional `pr\_number`, then dispatches the scenario-owned workflow. The wrapper is intentionally thin: scenario workflows still own their transport setup, credentials, VM class, expected oracle, and artifact manife... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/mantis.md | main | opebclaw | [
-0.07590852677822113,
0.00402945326641202,
-0.062446895986795425,
-0.05069839954376221,
0.01194411888718605,
-0.03344825655221939,
-0.035691022872924805,
0.020141569897532463,
0.006121251732110977,
0.03310254588723183,
-0.018973348662257195,
-0.050163187086582184,
0.001837562769651413,
0.0... | 0.151848 |
opens Telegram Desktop without configuring OpenClaw. - `--credential-source convex --credential-role ci` uses the shared credential broker instead of direct Telegram env tokens. Every PR-publishing scenario writes `mantis-evidence.json` next to its report. This schema is the handoff between scenario code and GitHub com... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/mantis.md | main | opebclaw | [
-0.049389101564884186,
0.00900209415704012,
-0.08996226638555527,
0.008452796377241611,
-0.02253131754696369,
-0.059478696435689926,
-0.07087722420692444,
0.04766061156988144,
0.07257844507694244,
0.09061121940612793,
0.02707722969353199,
-0.06566973775625229,
0.014896325767040253,
0.09338... | 0.10443 |
ref in the same VM. 11. Run the same scenario and capture candidate evidence. 12. Compare the oracle results and visual evidence. 13. Write Markdown, JSON, logs, screenshots, and optional trace artifacts. 14. Upload GitHub Actions artifacts. 15. Post a concise PR or Discord status message. The scenario should be able t... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/mantis.md | main | opebclaw | [
-0.08538811653852463,
-0.034795068204402924,
-0.011883336119353771,
-0.001814203686080873,
0.11961373686790466,
-0.03054782561957836,
0.010728695429861546,
0.06307055056095123,
-0.05598583072423935,
0.023101314902305603,
0.019460005685687065,
-0.043146148324012756,
0.05417333170771599,
-0.... | 0.030394 |
still need redaction discipline: private channel names, user names, or message content may appear. For public PRs, prefer GitHub Actions artifact links over inline images until the redaction story is stronger. ## Browser and VNC The browser lane has two modes: - \*\*Headless automation\*\*: default for CI. Chrome runs ... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/mantis.md | main | opebclaw | [
-0.006153101567178965,
-0.07349333167076111,
-0.05546051263809204,
-0.030950121581554413,
0.0806572437286377,
0.0037113113794475794,
-0.030900849029421806,
-0.01799602434039116,
-0.0191179271787405,
-0.030194103717803955,
0.06293366104364395,
-0.03955557197332382,
0.004082136787474155,
0.0... | 0.062252 |
on a generic QA automation PR. Raw logs, observed messages, and other bulky evidence stay in the Actions artifact. Production workflows should post those comments with the Mantis GitHub App, not with `github-actions[bot]`. Store the app id and private key as `MANTIS\_GITHUB\_APP\_ID` and `MANTIS\_GITHUB\_APP\_PRIVATE\_... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/mantis.md | main | opebclaw | [
-0.12557975947856903,
-0.028712565079331398,
-0.04092755541205406,
0.002641404280439019,
0.004891497548669577,
-0.007674697786569595,
0.008767715655267239,
0.002652017632499337,
0.06369800865650177,
0.07664552330970764,
0.056027188897132874,
-0.007968559861183167,
0.06862994283437729,
0.02... | 0.115715 |
of waiting for a maintainer command? - Should screenshots be redacted or cropped before upload for public PRs? | https://github.com/openclaw/openclaw/blob/main/docs/concepts/mantis.md | main | opebclaw | [
-0.06670516729354858,
0.06983955204486847,
-0.047575242817401886,
-0.007950610481202602,
0.13549934327602386,
-0.02091822773218155,
-0.022656423971056938,
0.012535423040390015,
-0.014479557983577251,
0.0348338708281517,
0.12660813331604004,
0.030770668759942055,
0.031042564660310745,
-0.03... | -0.006865 |
The \*\*OpenClaw App SDK\*\* is the public client API for apps outside the OpenClaw process. Use `@openclaw/sdk` when a script, dashboard, CI job, IDE extension, or other external app wants to connect to the Gateway, start agent runs, stream events, wait for results, cancel work, or inspect Gateway resources. The App S... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/openclaw-sdk.md | main | opebclaw | [
-0.08497060835361481,
0.011802646331489086,
-0.05962173640727997,
0.029026279225945473,
0.04099935293197632,
-0.10444346815347672,
-0.04817093908786774,
0.023849202319979668,
0.05320040509104729,
-0.027442120015621185,
0.030875759199261665,
-0.05509646236896515,
-0.006567188072949648,
-0.0... | 0.092515 |
oc.agents.get("main"); const run = await agent.run({ input: "Review this pull request and suggest the smallest safe fix.", model: "openai/gpt-5.5", sessionKey: "main", timeoutMs: 30\_000, }); for await (const event of run.events()) { const data = event.data as { delta?: unknown }; if (event.type === "assistant.delta" &... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/openclaw-sdk.md | main | opebclaw | [
-0.0012973059201613069,
0.05394480749964714,
-0.10708191990852356,
0.10176494717597961,
0.029107963666319847,
-0.06084038317203522,
-0.030767910182476044,
0.06711476296186447,
0.0863238275051117,
-0.051582325249910355,
-0.008618335239589214,
-0.01840854622423649,
-0.0642264187335968,
-0.06... | 0.105614 |
Each call requires one explicit `sessionKey`, `runId`, or `taskId` scope: ```typescript const { artifacts } = await oc.artifacts.list({ sessionKey: "main" }); const first = artifacts[0]; if (first) { const { artifact } = await oc.artifacts.get(first.id, { sessionKey: "main" }); const download = await oc.artifacts.downl... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/openclaw-sdk.md | main | opebclaw | [
-0.09175744652748108,
0.06340274214744568,
-0.03308864310383797,
0.03975982964038849,
0.043694186955690384,
-0.07643595337867737,
0.08219307661056519,
0.07592152804136276,
0.04229482635855675,
0.05436137691140175,
-0.0403829850256443,
-0.04764224961400032,
-0.08989454805850983,
0.056560821... | 0.104152 |
Parallel specialist lanes let one Gateway route different chats or rooms to different agents, while keeping the user experience fast. The trick is to treat parallelism as a scarce-resource design problem, not just as "more agents". ## First principles A specialist lane only improves throughput when it reduces contentio... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/parallel-specialist-lanes.md | main | opebclaw | [
-0.012981441803276539,
-0.0294797345995903,
-0.07357125729322433,
-0.026816392317414284,
-0.09296289831399918,
-0.08970194309949875,
-0.000739650393370539,
-0.037961047142744064,
0.005899342242628336,
0.0386868491768837,
-0.02668038383126259,
0.008686541579663754,
-0.012740311212837696,
0.... | 0.11808 |
OpenClaw "presence" is a lightweight, best-effort view of: - the \*\*Gateway\*\* itself, and - \*\*clients connected to the Gateway\*\* (mac app, WebChat, CLI, etc.) Presence is used primarily to render the macOS app's \*\*Instances\*\* tab and to provide quick operator visibility. ## Presence fields (what shows up) Pr... | https://github.com/openclaw/openclaw/blob/main/docs/concepts/presence.md | main | opebclaw | [
-0.04622180759906769,
-0.01619575172662735,
-0.06351853162050247,
0.012833724729716778,
0.050286971032619476,
-0.12901023030281067,
0.015115451999008656,
-0.009374236688017845,
0.03772066533565521,
-0.005822357255965471,
0.07338724285364151,
-0.05118235573172569,
0.002199191600084305,
-0.0... | 0.193559 |
Diagnostics flags let you enable targeted debug logs without turning on verbose logging everywhere. Flags are opt-in and have no effect unless a subsystem checks them. ## How it works - Flags are strings (case-insensitive). - You can enable flags in config or via an env override. - Wildcards are supported: - `telegram.... | https://github.com/openclaw/openclaw/blob/main/docs/diagnostics/flags.md | main | opebclaw | [
0.040551524609327316,
0.058908529579639435,
-0.05129793658852577,
0.03473309800028801,
0.06231346353888512,
-0.13353854417800903,
0.0076461960561573505,
-0.009378130547702312,
-0.015112742781639099,
0.02871273085474968,
0.0069041126407682896,
-0.02842160500586033,
-0.008844386786222458,
0.... | 0.083067 |
volume for the specific subsystem. - Use [/logging](/logging) to change log destinations, levels, and redaction. ## Related - [Gateway diagnostics](/gateway/diagnostics) - [Gateway troubleshooting](/gateway/troubleshooting) | https://github.com/openclaw/openclaw/blob/main/docs/diagnostics/flags.md | main | opebclaw | [
0.07310042530298233,
-0.011546831578016281,
-0.04084518179297447,
0.0011607696069404483,
0.009250369854271412,
-0.06815961748361588,
-0.045038189738988876,
0.0545525848865509,
-0.009256432764232159,
-0.022596171125769615,
-0.02629305236041546,
-0.004761777818202972,
-0.004173163324594498,
... | 0.06662 |
Cron is the Gateway's built-in scheduler. It persists jobs, wakes the agent at the right time, and can deliver output back to a chat channel or webhook endpoint. ## Quick start ```bash openclaw cron create "2026-02-01T16:00:00Z" \ --name "Reminder" \ --session main \ --system-event "Reminder: check the cron docs draft"... | https://github.com/openclaw/openclaw/blob/main/docs/automation/cron-jobs.md | main | opebclaw | [
-0.07625596970319748,
-0.03946198523044586,
-0.11248553544282913,
0.04121652990579605,
0.01163790374994278,
-0.05132988095283508,
-0.03112105280160904,
-0.019920967519283295,
0.033395830541849136,
-0.051976174116134644,
-0.035641346126794815,
-0.014329980127513409,
-0.03138796612620354,
-0... | 0.106028 |
of waiting for the full job budget. - If you use system cron or another external scheduler to run `openclaw agent`, wrap it with a hard-kill escalation even though the CLI handles `SIGTERM`/`SIGINT`. Gateway-backed runs ask the Gateway to abort accepted runs; local and embedded fallback runs receive the same abort sign... | https://github.com/openclaw/openclaw/blob/main/docs/automation/cron-jobs.md | main | opebclaw | [
-0.07787036895751953,
-0.03719986602663994,
-0.064480260014534,
0.03572617098689079,
0.006159507669508457,
-0.04508846625685692,
-0.0828251764178276,
-0.016561198979616165,
0.028672441840171814,
-0.027337923645973206,
0.0030426650773733854,
0.036988090723752975,
-0.01641579158604145,
-0.03... | 0.060869 |
optionally wake the heartbeat (`--wake now` or `--wake next-heartbeat`). They can use the target main session's last delivery context for replies, but they do not append routine cron turns to the human chat lane and do not extend daily/idle reset freshness for the target session. \*\*Isolated\*\* jobs run a dedicated a... | https://github.com/openclaw/openclaw/blob/main/docs/automation/cron-jobs.md | main | opebclaw | [
-0.0847368985414505,
-0.013770271092653275,
-0.06021667644381523,
0.021418064832687378,
0.048729412257671356,
-0.04308157041668892,
0.026200242340564728,
-0.044164784252643585,
0.012477723881602287,
-0.08225798606872559,
-0.07696717232465744,
-0.014934630133211613,
-0.03209726884961128,
0.... | 0.071411 |
model override 4. Agent/default model selection Fast mode follows the resolved live selection too. If the selected model config has `params.fastMode`, isolated cron uses that by default. A stored session `fastMode` override still wins over config in either direction. If an isolated run hits a live model-switch handoff,... | https://github.com/openclaw/openclaw/blob/main/docs/automation/cron-jobs.md | main | opebclaw | [
-0.026950791478157043,
-0.05150097981095314,
-0.03820610046386719,
0.07420515269041061,
0.06275467574596405,
-0.06416762620210648,
-0.05268321931362152,
-0.01438287552446127,
0.004707308020442724,
-0.054848812520504,
0.08878887444734573,
0.008791886270046234,
-0.03334910050034523,
-0.00655... | -0.000395 |
infer a reply language from channel, locale, or previous messages. Put the language rule in the scheduled message or template: ```bash openclaw cron edit \ --message "Summarize the updates. Respond in Chinese; keep URLs, code, and product names unchanged." ``` For template files, keep the language instruction in the re... | https://github.com/openclaw/openclaw/blob/main/docs/automation/cron-jobs.md | main | opebclaw | [
-0.035976480692625046,
0.014851152896881104,
-0.0033575515262782574,
0.016798537224531174,
0.030883148312568665,
0.002095191041007638,
0.07330776005983353,
-0.009070106782019138,
0.014135428704321384,
-0.07463587075471878,
-0.012156764045357704,
-0.10709438472986221,
0.03448240086436272,
0... | 0.045098 |
Wizard setup (recommended) ```bash openclaw webhooks gmail setup --account openclaw@gmail.com ``` This writes `hooks.gmail` config, enables the Gmail preset, and uses Tailscale Funnel for the push endpoint. ### Gateway auto-start When `hooks.enabled=true` and `hooks.gmail.account` is set, the Gateway starts `gog gmail ... | https://github.com/openclaw/openclaw/blob/main/docs/automation/cron-jobs.md | main | opebclaw | [
-0.0704292356967926,
-0.0107900220900774,
-0.00567854568362236,
-0.021895475685596466,
-0.03694571927189827,
-0.05091932788491249,
0.018050633370876312,
-0.027242228388786316,
0.009854095056653023,
-0.017124608159065247,
-0.0011132402578368783,
-0.021722180768847466,
0.0238510649651289,
-0... | -0.056801 |
and defaults to 8. Isolated cron agent turns use the queue's dedicated `cron-nested` execution lane internally, so raising this value lets independent cron LLM runs progress in parallel instead of only starting their outer cron wrappers. The shared non-cron `nested` lane is not widened by this setting. `cron.store` is ... | https://github.com/openclaw/openclaw/blob/main/docs/automation/cron-jobs.md | main | opebclaw | [
-0.08691658079624176,
-0.049010179936885834,
-0.1131371483206749,
-0.007669507991522551,
-0.0349276177585125,
-0.06022217124700546,
-0.04486161097884178,
0.03138680383563042,
-0.04336027428507805,
-0.022817296907305717,
0.059719834476709366,
-0.015916550531983376,
-0.01752806082367897,
-0.... | 0.008369 |
Task Flow is the flow orchestration substrate that sits above [background tasks](/automation/tasks). It manages durable multi-step flows with their own state, revision tracking, and sync semantics while individual tasks remain the unit of detached work. ## When to use Task Flow Use Task Flow when work spans multiple se... | https://github.com/openclaw/openclaw/blob/main/docs/automation/taskflow.md | main | opebclaw | [
-0.10729333013296127,
-0.034535184502601624,
-0.033715780824422836,
-0.01570100523531437,
0.005221656057983637,
-0.00023773760767653584,
-0.024940891191363335,
-0.03272133320569992,
0.04631486535072327,
0.0077062202617526054,
-0.06480038911104202,
-0.007901610806584358,
0.022579651325941086,... | 0.188319 |
(3) delivers it. Task Flow creates each step as a background task, waits for completion, then moves to the next step. ``` Flow: weekly-report Step 1: gather-data → task created → succeeded Step 2: generate-report → task created → succeeded Step 3: deliver → task created → running ``` ### Mirrored mode Task Flow observe... | https://github.com/openclaw/openclaw/blob/main/docs/automation/taskflow.md | main | opebclaw | [
-0.1018093079328537,
-0.05865519121289253,
-0.04488198831677437,
-0.017105409875512123,
0.024427417665719986,
-0.036892980337142944,
-0.061852265149354935,
-0.0677918940782547,
0.021071841940283775,
0.007364922668784857,
-0.06686726212501526,
0.035961177200078964,
-0.06107010692358017,
-0.... | 0.151742 |
Looking for scheduling? See [Automation](/automation) for choosing the right mechanism. This page is the activity ledger for background work, not the scheduler. Background tasks track work that runs \*\*outside your main conversation session\*\*: ACP runs, subagent spawns, isolated cron job executions, and CLI-initiate... | https://github.com/openclaw/openclaw/blob/main/docs/automation/tasks.md | main | opebclaw | [
-0.057101599872112274,
-0.018163789063692093,
-0.02060556225478649,
-0.019103553146123886,
0.01795872673392296,
-0.0938885360956192,
0.012257169000804424,
-0.038922667503356934,
0.0675930306315422,
-0.009329620748758316,
-0.038994647562503815,
-0.05585445091128349,
-0.029749879613518715,
0... | 0.153197 |
own session. Session-backed `image\_generate`, `music\_generate`, and `video\_generate` runs also use `silent` notify policy. They still create task records, but completion is handed back to the original agent session as an internal wake so the agent can write the follow-up message and attach the finished media itself.... | https://github.com/openclaw/openclaw/blob/main/docs/automation/tasks.md | main | opebclaw | [
-0.026354502886533737,
-0.03360135108232498,
0.002458056202158332,
0.04484452307224274,
0.07842400670051575,
-0.07128652930259705,
0.03727143257856369,
-0.05385592207312584,
0.11398570239543915,
0.019927728921175003,
0.0008107951725833118,
0.005423491355031729,
0.027243033051490784,
-0.019... | 0.106891 |
a task reaches a terminal state, OpenClaw notifies you. There are two delivery paths: \*\*Direct delivery\*\* - if the task has a channel target (the `requesterOrigin`), the completion message goes straight to that channel (Telegram, Discord, Slack, etc.). Group and channel task completions are instead routed through t... | https://github.com/openclaw/openclaw/blob/main/docs/automation/tasks.md | main | opebclaw | [
-0.020525308325886726,
-0.0677027478814125,
-0.014990551397204399,
0.051474083214998245,
0.007157070096582174,
-0.11247765272855759,
-0.00030493727535940707,
-0.06503280997276306,
0.1096864566206932,
0.015415986999869347,
-0.011331590823829174,
-0.014138462953269482,
-0.020507510751485825,
... | 0.106622 |
job, then recover terminal status from persisted cron run logs/job state before falling back to `lost`. Only the Gateway process is authoritative for the in-memory cron active-job set; offline CLI audit uses durable history but does not mark a cron task lost solely because that local Set is empty. - CLI tasks with run ... | https://github.com/openclaw/openclaw/blob/main/docs/automation/tasks.md | main | opebclaw | [
-0.06021339073777199,
-0.004293452948331833,
-0.021341044455766678,
0.002777854213491082,
0.07921736687421799,
-0.012030585668981075,
0.0020599314011633396,
-0.03560039773583412,
0.0009229090064764023,
-0.007406503893435001,
0.0018284766701981425,
-0.005781702697277069,
-0.043513406068086624... | 0.03845 |
7 days). During retention, lost tasks still appear in audit as warnings; after `cleanupAfter` expires or when cleanup metadata is missing, they are errors. Deletes records past their `cleanupAfter` date. \*\*Retention:\*\* terminal task records are kept for \*\*7 days\*\*, then automatically pruned. No configuration ne... | https://github.com/openclaw/openclaw/blob/main/docs/automation/tasks.md | main | opebclaw | [
-0.04685633257031441,
-0.027957556769251823,
-0.03673451021313667,
0.05089772492647171,
0.06351129710674286,
-0.045741043984889984,
-0.014505011960864067,
-0.05803748220205307,
0.010852186009287834,
-0.018273860216140747,
-0.01054688636213541,
0.0004380272584967315,
0.01785491779446602,
0.... | 0.090997 |
Standing orders grant your agent \*\*permanent operating authority\*\* for defined programs. Instead of giving individual task instructions each time, you define programs with clear scope, triggers, and escalation rules - and the agent executes autonomously within those boundaries. This is the difference between tellin... | https://github.com/openclaw/openclaw/blob/main/docs/automation/standing-orders.md | main | opebclaw | [
-0.06979398429393768,
0.061190180480480194,
-0.04313124343752861,
0.02314889430999756,
-0.0567738339304924,
0.025029730051755905,
-0.053401269018650055,
-0.002031759824603796,
-0.012592840008437634,
0.028647473081946373,
-0.012628850527107716,
0.053818121552467346,
0.03817303106188774,
0.0... | 0.083484 |
unknowns." ``` ## Examples ### Example 1: content and social media (weekly cycle) ```markdown ## Program: Content & Social Media \*\*Authority:\*\* Draft content, schedule posts, compile engagement reports \*\*Approval gate:\*\* All posts require owner review for first 30 days, then standing approval \*\*Trigger:\*\* W... | https://github.com/openclaw/openclaw/blob/main/docs/automation/standing-orders.md | main | opebclaw | [
-0.076434426009655,
0.013352971524000168,
-0.05915364250540733,
0.03650563955307007,
0.08460688591003418,
0.017351673915982246,
0.026547757908701897,
-0.022281968966126442,
-0.06044375151395798,
0.019640600308775902,
-0.006593522150069475,
-0.015338943339884281,
0.024263786152005196,
-0.06... | 0.178131 |
Rules (All Programs) - [Common escalation criteria] - [Approval gates that apply across programs] ``` Each program should have: - Its own \*\*trigger cadence\*\* (weekly, monthly, event-driven, continuous) - Its own \*\*approval gates\*\* (some programs need more oversight than others) - Clear \*\*boundaries\*\* (the a... | https://github.com/openclaw/openclaw/blob/main/docs/automation/standing-orders.md | main | opebclaw | [
-0.0071731493808329105,
-0.008505446836352348,
-0.008375346660614014,
-0.026155132800340652,
-0.006041370332241058,
-0.030399063602089882,
-0.007853712886571884,
0.006877991370856762,
-0.07560954242944717,
0.016391975805163383,
-0.013686045072972775,
0.006839510053396225,
0.01085766870528459... | 0.116212 |
OpenClaw runs work in the background through tasks, scheduled jobs, inferred commitments, event hooks, and standing instructions. This page helps you choose the right mechanism and understand how they fit together. ## Quick decision guide ```mermaid flowchart TD START([What do you need?]) --> Q1{Schedule work?} START -... | https://github.com/openclaw/openclaw/blob/main/docs/automation/index.md | main | opebclaw | [
-0.028263747692108154,
0.009105556644499302,
-0.03987336531281471,
-0.0015551748219877481,
0.007764250971376896,
0.008789438754320145,
-0.049251146614551544,
-0.00883068609982729,
0.028986210003495216,
-0.051013439893722534,
-0.023013340309262276,
-0.05626780167222023,
-0.03197663277387619,
... | 0.071401 |
are opt-in, short-lived follow-up memories. OpenClaw infers them from normal conversations, scopes them to the same agent and channel, and delivers due check-ins through heartbeat. Exact user-requested reminders still belong to cron. See [Inferred Commitments](/concepts/commitments). ### Task Flow Task Flow is the flow... | https://github.com/openclaw/openclaw/blob/main/docs/automation/index.md | main | opebclaw | [
-0.09826242923736572,
0.014388003386557102,
-0.08152075111865997,
0.011994925327599049,
0.03418829292058945,
-0.03741747885942459,
-0.011580056510865688,
-0.02318972535431385,
0.06022880598902702,
-0.016452496871352196,
0.0023637828417122364,
0.053335268050432205,
-0.014179842546582222,
-0... | 0.205826 |
Hooks are small scripts that run when something happens inside the Gateway. They can be discovered from directories and inspected with `openclaw hooks`. The Gateway loads internal hooks only after you enable hooks or configure at least one hook entry, hook pack, legacy handler, or extra hook directory. There are two ki... | https://github.com/openclaw/openclaw/blob/main/docs/automation/hooks.md | main | opebclaw | [
-0.06028498709201813,
-0.0054407063871622086,
-0.05566410720348358,
0.02044605277478695,
0.03473430499434471,
-0.08120636641979218,
-0.010506043210625648,
-0.027920981869101524,
0.01130127627402544,
-0.02446531131863594,
0.018852725625038147,
-0.03668573871254921,
-0.034685518592596054,
-0... | 0.117006 |
(e.g., `["darwin", "linux"]`) | | `requires` | Required `bins`, `anyBins`, `env`, or `config` paths | | `always` | Bypass eligibility checks (boolean) | | `install` | Installation methods | ### Handler implementation ```typescript const handler = async (event) => { if (event.type !== "command" || event.action !== "new"... | https://github.com/openclaw/openclaw/blob/main/docs/automation/hooks.md | main | opebclaw | [
-0.03704209625720978,
-0.06909523159265518,
0.03653377294540405,
0.06075788661837578,
0.0511036179959774,
-0.0772404819726944,
0.08522837609052658,
0.09077965468168259,
0.00523117883130908,
0.05724551901221275,
-0.04880736395716667,
-0.0727623775601387,
-0.057643402367830276,
0.05287866666... | 0.085146 |
override precedence: 1. \*\*Bundled hooks\*\*: shipped with OpenClaw 2. \*\*Plugin hooks\*\*: hooks bundled inside installed plugins 3. \*\*Managed hooks\*\*: `~/.openclaw/hooks/` (user-installed, shared across workspaces). Extra directories from `hooks.internal.load.extraDirs` share this precedence. 4. \*\*Workspace h... | https://github.com/openclaw/openclaw/blob/main/docs/automation/hooks.md | main | opebclaw | [
-0.04456199333071709,
-0.08026669919490814,
-0.0019483764190226793,
0.01952475681900978,
-0.025859015062451363,
-0.05587751045823097,
-0.031293634325265884,
-0.0614793486893177,
-0.024090800434350967,
-0.026272591203451157,
0.023042678833007812,
0.012901009060442448,
-0.031048230826854706,
... | 0.010162 |
"value" } } } } } } ``` Extra hook directories: ```json { "hooks": { "internal": { "load": { "extraDirs": ["/path/to/more/hooks"] } } } } ``` The legacy `hooks.internal.handlers` array config format is still supported for backwards compatibility, but new hooks should use the discovery-based system. ## CLI reference ```... | https://github.com/openclaw/openclaw/blob/main/docs/automation/hooks.md | main | opebclaw | [
-0.049680985510349274,
0.019975416362285614,
-0.06694135814905167,
0.022046789526939392,
-0.004495880100876093,
-0.050610046833753586,
0.0008788102422840893,
0.014162341132760048,
-0.11034227162599564,
-0.03889703378081322,
0.032749298959970474,
-0.024776771664619446,
-0.02552013471722603,
... | 0.143295 |
# Tweakcn Custom Theme Import Design Status: approved in terminal on 2026-04-22 ## Summary Add exactly one browser-local custom Control UI theme slot that can be imported from a tweakcn share link. The existing built-in theme families remain `claw`, `knot`, and `dash`. The new `custom` family behaves like a normal Open... | https://github.com/openclaw/openclaw/blob/main/docs/superpowers/specs/2026-04-22-tweakcn-custom-theme-import-design.md | main | opebclaw | [
-0.08178545534610748,
-0.009093433618545532,
-0.020590705797076225,
0.061195358633995056,
0.02993546985089779,
-0.02920067310333252,
0.01285142358392477,
-0.0161846112459898,
-0.026673998683691025,
-0.013979361392557621,
0.0041588060557842255,
0.03134297952055931,
-0.01884348876774311,
-0.... | 0.02465 |
sourceUrl: string; themeId: string; label: string; importedAt: string; light: Record<string, string>; dark: Record<string, string>; }; ``` Notes: - `sourceUrl` stores the original user input after normalization. - `themeId` is the tweakcn theme id extracted from the URL. - `label` is the tweakcn `name` field when prese... | https://github.com/openclaw/openclaw/blob/main/docs/superpowers/specs/2026-04-22-tweakcn-custom-theme-import-design.md | main | opebclaw | [
-0.07045022398233414,
0.05834789201617241,
-0.05552644282579422,
0.0789402648806572,
0.024127840995788574,
-0.01971203275024891,
0.008512100204825401,
0.006565673276782036,
0.0003451660741120577,
-0.05109203979372978,
-0.017820725217461586,
-0.023261578753590584,
0.002783596282824874,
-0.0... | 0.090926 |
block overrides `font-sans`, `font-mono`, or `radius`, the mode-local value wins. ### Tokens derived for OpenClaw The importer derives OpenClaw-only variables from the imported base colors: - `--bg-accent` - `--bg-elevated` - `--bg-hover` - `--panel` - `--panel-strong` - `--panel-hover` - `--chrome` - `--chrome-strong`... | https://github.com/openclaw/openclaw/blob/main/docs/superpowers/specs/2026-04-22-tweakcn-custom-theme-import-design.md | main | opebclaw | [
-0.12610730528831482,
-0.055091213434934616,
-0.031133612617850304,
0.062320776283741,
0.06042438745498657,
-0.05033765360713005,
0.0036419134121388197,
0.012968456372618675,
-0.06912267953157425,
-0.030101418495178223,
0.0604046955704689,
-0.049550626426935196,
-0.00319097307510674,
-0.00... | 0.070981 |
## Goal Move the application toward a cleaner, faster, more maintainable product without breaking current workflows or hiding risk in broad refactors. The work should land as small, reviewable slices with proof for each touched surface. ## Principles - Preserve current architecture unless a boundary is demonstrably cau... | https://github.com/openclaw/openclaw/blob/main/docs/reference/application-modernization-plan.md | main | opebclaw | [
-0.09100276231765747,
-0.0006415102398023009,
0.03427036106586456,
0.000808146083727479,
0.03723976016044617,
-0.028446247801184654,
-0.023499896749854088,
0.07804384082555771,
-0.11387369781732559,
0.04767575487494469,
-0.039350755512714386,
0.0063786315731704235,
-0.029707029461860657,
0... | 0.102891 |
Phase 5: Type, contract, and test hardening Raise correctness at the boundary points users and plugin authors depend on. - Replace loose runtime strings with discriminated unions or closed code lists. - Validate external inputs with existing schema helpers or zod. - Add contract tests around plugin manifests, provider ... | https://github.com/openclaw/openclaw/blob/main/docs/reference/application-modernization-plan.md | main | opebclaw | [
-0.08996633440256119,
-0.031154919415712357,
0.030916983261704445,
-0.02601887285709381,
0.05417285114526749,
-0.039596714079380035,
-0.0726182609796524,
0.04970371350646019,
-0.14338859915733337,
0.08583234250545502,
0.016309278085827827,
0.01626964844763279,
-0.058765992522239685,
0.0321... | 0.059678 |
Cards should be used for repeated items, modals, and framed tools, not for every page section. - Avoid one-note color palettes and decorative backgrounds that compete with operational content. - Dense product surfaces should optimize for scanning, comparison, and repeated use. ## Handoff format Report: - What changed. ... | https://github.com/openclaw/openclaw/blob/main/docs/reference/application-modernization-plan.md | main | opebclaw | [
-0.08887496590614319,
0.09872862696647644,
0.004358601290732622,
0.011172584258019924,
0.08115774393081665,
0.053001273423433304,
0.01805982179939747,
-0.005529870744794607,
-0.07540418207645416,
-0.08055965602397919,
-0.01634281873703003,
0.14131246507167816,
-0.011961604468524456,
0.0215... | 0.074387 |
# Secret placeholder conventions Use placeholders that are human-readable but do not resemble real secrets. ## Recommended style - Prefer descriptive values like `example-openai-key-not-real` or `example-discord-bot-token`. - For shell snippets, prefer `${OPENAI\_API\_KEY}` over inline token-like strings. - Keep exampl... | https://github.com/openclaw/openclaw/blob/main/docs/reference/secret-placeholder-conventions.md | main | opebclaw | [
-0.03661949187517166,
0.04802520200610161,
-0.04410776495933533,
0.019279928877949715,
0.08038363605737686,
-0.05336499586701393,
0.04859043285250664,
0.062279049307107925,
0.07399623841047287,
0.008384314365684986,
0.01948116160929203,
-0.057091113179922104,
0.0825522243976593,
-0.0118493... | 0.109136 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.