@xiaozhiclaw/qlogicagent-core 0.6.6 → 0.6.8

package/README.md

@@ -16,6 +16,24 @@ persistence. Login, permissions, conversations, business commits, audit, files,
 database connections and workers remain with the product. Model routing,
 credentials and accounting remain with its model service.
 
+Optional `executeWithModelReconnects(runtime, context, id, options)` drives the
+same durable runtime. The default allows five reconnects after the first rejected
+attempt, with exponential backoff anchored to the persisted attempt timestamps.
+The original owner must prove retry safety through `authorizeRecovery`; unknown
+results remain pending reconciliation. Cancellation, authorization, task inputs,
+receipts and budgets retain their existing owners. Plain `runtime.execute` is
+unchanged. An explicit `operatorRetryRequestId` permits only one additional retry
+of that exact original rejection; it never resets the automatic allowance.
+
+`AutomationSchedule`, `validAutomationSchedule` and `automationOccurrence` provide
+optional once, interval, daily, weekly, monthly and yearly trigger calculation.
+Pass a fixed `utcOffsetMinutes` explicitly; IANA zones and daylight-saving
+transitions are not resolved by this API. Month ends clamp to the last valid day.
+The result is the latest due occurrence, not a backlog of missed jobs. The host
+chooses catch-up/overlap policy and atomically registers a scope-bound occurrence
+and its original task/input in its own store. Business report periods stay in the
+product. These exports require no configuration package or model-service SDK.
+
 ```ts
 import { createCloudAgent } from "@xiaozhiclaw/qlogicagent-core";
 
@@ -200,8 +218,18 @@ with a different context projection requires explicit product migration; Core
 does not reinterpret its saved input or tool receipts. Checkpoints without an
 assistant context retain their existing binding protocol.
 Optional `goal` requires a product `verifyGoal` callback before semantic completion.
-Products may resolve a user-confirmed goal from receipts with `resolveGoal`, and
-return independent verification evidence to the same loop through `goalFeedback`.
+Products may resolve a user-confirmed goal from receipts with `resolveGoal`.
+Incomplete evidence returns to the same loop by default; `goalFeedback` can
+override its wording or return `[]` to stop. Unchanged incomplete evidence stops
+repeated completion claims. Additional read receipts or worker replacement do not
+prove business progress; this comparison reuses the persisted outcome.
+Two consecutive identical batches rejected before tool dispatch, with no new
+input, end with `TOOL_ARGUMENTS_STALLED`. The check uses checkpoint history;
+changed proposals, mixed valid work and new input keep their normal execution path.
+CloudTask terminal events report `taskOutcome: "satisfied" | "incomplete"` after
+independent verification; an incomplete outcome also has `turnStatus: "stopped"`.
+Without a confirmed goal, `taskOutcome` remains `"unverified"`; lifecycle completion
+does not prove a business result. Product budgets, authorization and waits still apply.
 The model can correct its work or use existing input/confirmation wait tools;
 the product continues to own the acceptance criteria and verification rules.
 Criteria may carry opaque product `parameters`; verification can return `expected`,
@@ -211,7 +239,7 @@ Context anchors carry explicit revision/supersession/source references through
 compaction, keeping unresolved conflicts visible. Product tools retrieve material.
 Optional `runtime.prepareContext(messages, context)` asynchronously retrieves and
 projects product task materials before inference. `context` carries the model,
-input token budget, session/turn IDs and cancellation signal.
+input token budget, resident tool schemas, session/turn IDs and cancellation signal.
 The budget is `Infinity` when context-capacity metadata is explicitly disabled;
 retrieval still runs and the provider remains responsible for its actual capacity.
 Return `{ messages, droppedCount, strategy }`; retrieval and fact updates are applied even
@@ -224,8 +252,22 @@ fills remaining capacity with recent history. These essential messages may excee
 capacity; Core does not silently drop them or pretend an estimate proves model fit.
 Optional `runtime.countMessageTokens(message, model)` supplies model-aware counts
 for the default window. Its fallback is an advisory serialized-message estimate.
-The input budget reserves configured maximum output tokens rather than a fixed
-percentage; provider context errors remain authoritative.
+The default projection reserves configured maximum output tokens and resident
+tool schema cost before selecting message history; provider context errors remain authoritative.
+Optional `runtime.summarizeContext({previousSummary, messages}, context)` summarizes
+only omitted complete protocol groups and merges the prior summary. Configure it
+through the product's existing metered model service. The callback receives a
+summary token allowance and cancellation signal; it owns model authorization,
+cost accounting, deduplication and unknown-outcome recovery for that work. Its
+usage is not automatically included in the main loop's model-operation ledger.
+The default selector reserves a quarter of message capacity for a summary, pins
+original intent, the latest amendment, facts and the latest complete protocol group,
+and passes remaining older groups to the callback. It saves one derived summary in
+the same checkpoint. Empty/oversized summaries or callback failures fail explicitly;
+they do not silently fall back to discarding history. A provider-acknowledged response
+replays its original projection without invoking the summarizer again.
+`prepareContext`, when supplied, owns the whole projection and takes precedence.
+Summaries are data, never instructions, authenticated facts or authorization.
 After retrieving current product facts, call public
 `prepareSlidingWindowContext(messages, context, countMessageTokens?)` to reuse
 the same canonical window implementation. Product retrieval must not accidentally
@@ -245,6 +287,15 @@ that contract and enforce the original ceiling atomically; it remains responsibl
 for supported currencies, precision and settlement. The model transport supplies
 normalized usage totals; source names in usage records stay opaque to Core.
 
+Products explicitly mark independent tools with `meta.executionMode: "parallel"`
+and set `config.maxConcurrentTools` to permit overlap. Omitted or `"sequential"`
+execution modes form barriers: all earlier calls settle before that call starts,
+and later calls wait for it. Consecutive parallel-safe calls share the same bounded
+executor. No safety is inferred from tool names, categories or model arguments.
+Results stream as each call finishes; model/checkpoint history retains original
+call order. Cancellation or a wait stops queued calls; already dispatched effects
+still need their original receipts and the existing reconciliation path.
+
 Cloud execution does not invent a round limit, per-response tool-call limit,
 retry-count limit, total presentation-block limit, compaction-count stop or default
 tool timeout. Products may supply per-tool `meta.executionTimeoutMs`; task deadlines

package/dist/automation.js

@@ -0,0 +1 @@
+function s(e){return typeof e=="string"&&/^\d{4}-\d\d-\d\dT\d\d:\d\d$/u.test(e)&&Number(e.slice(0,4))>=2e3&&Number.isFinite(Date.parse(e+":00Z"))&&new Date(e+":00Z").toISOString().slice(0,16)===e}function D(e){if(!e||typeof e!="object"||Array.isArray(e))return!1;let t=e,y=(...i)=>Object.keys(t).every(m=>i.includes(m)),r=(i,m,o)=>Number.isInteger(i)&&Number(i)>=m&&Number(i)<=o;return t.frequency==="once"?y("frequency","at")&&s(t.at):t.frequency==="interval"?y("frequency","every","unit","startAt")&&["minutes","hours"].includes(String(t.unit))&&r(t.every,1,t.unit==="minutes"?10080:168)&&s(t.startAt):y("frequency","time","weekday","monthDay","month")&&typeof t.time=="string"&&/^([01]\d|2[0-3]):[0-5]\d$/u.test(t.time)&&(t.frequency==="daily"&&t.weekday===void 0&&t.monthDay===void 0&&t.month===void 0||t.frequency==="weekly"&&r(t.weekday,1,7)&&t.monthDay===void 0&&t.month===void 0||t.frequency==="monthly"&&r(t.monthDay,1,31)&&t.weekday===void 0&&t.month===void 0||t.frequency==="yearly"&&r(t.month,1,12)&&r(t.monthDay,1,31)&&t.weekday===void 0)}function T(e,t,y){if(!D(e)||!Number.isFinite(t.getTime())||!Number.isInteger(y)||Math.abs(y)>840)throw Error("CLOUD_AUTOMATION_SCHEDULE_INVALID");let r=y*6e4;if(e.frequency==="once"||e.frequency==="interval"){let u=Date.parse((e.frequency==="once"?e.at:e.startAt)+":00Z")-r;if(t.getTime()<u)return null;let a=e.frequency==="interval"?e.every*(e.unit==="minutes"?6e4:36e5):0,f=u+(a?Math.floor((t.getTime()-u)/a)*a:0);return{scheduledAt:new Date(f).toISOString(),scheduledDay:new Date(f+r).toISOString().slice(0,10)}}let i=new Date(t.getTime()+r),m=(u,a,f)=>new Date(Date.UTC(u,a,f)),o=m(i.getUTCFullYear(),i.getUTCMonth(),i.getUTCDate()),[g,l]=e.time.split(":").map(Number),d=g*36e5+l*6e4,c=(u,a)=>m(u,a,Math.min(e.monthDay,m(u,a+1,0).getUTCDate())),n=new Date(o);return e.frequency==="yearly"&&(n=c(o.getUTCFullYear(),e.month-1)),e.frequency==="monthly"&&(n=c(o.getUTCFullYear(),o.getUTCMonth())),e.frequency==="weekly"&&n.setUTCDate(n.getUTCDate()-((n.getUTCDay()||7)-e.weekday+7)%7),n.getTime()+d>i.getTime()&&(e.frequency==="yearly"?n=c(o.getUTCFullYear()-1,e.month-1):e.frequency==="monthly"?n=c(o.getUTCFullYear(),o.getUTCMonth()-1):n.setUTCDate(n.getUTCDate()-(e.frequency==="weekly"?7:1))),{scheduledAt:new Date(n.getTime()+d-r).toISOString(),scheduledDay:n.toISOString().slice(0,10)}}export{T as automationOccurrence,D as validAutomationSchedule};