eve 0.61.0 → 0.62.0

package/dist/src/setup/scaffold/index.js

@@ -1 +1 @@
-import{WriteFileExistsError}from"./files.js";import{CURRENT_DIRECTORY_PROJECT_NAME,byokProviderEnvVar,isEveProject,scaffoldBaseProject}from"./create/project.js";import{scaffoldExtensionProject}from"./create/extension.js";import{DEFAULT_SLACK_CONNECTOR_SLUG,SLACK_CHANNEL_DEFAULT_ROUTE,deriveSlackConnectorSlug,ensureChannel,hasVercelHostFramework,isNextJsProject,listAuthoredChannels,normalizeSlackConnectorSlug,resolveVercelHostFrameworkPreset}from"./update/channels.js";import{SCAFFOLDABLE_CHANNELS}from"./channels-catalog.js";import{CONNECTION_CATALOG,CUSTOM_CONNECTION_SLUG,SUPPORTED_PROTOCOLS,catalogSlugs,effectiveProtocols,endpointForProtocol,getCatalogEntry,isValidConnectionSlug}from"./connections/catalog.js";import{ensureConnection,ensureConnectionDependencies,listAuthoredConnections}from"./update/connections.js";import{HumanActionRequiredError}from"../human-action.js";export{CONNECTION_CATALOG,CURRENT_DIRECTORY_PROJECT_NAME,CUSTOM_CONNECTION_SLUG,DEFAULT_SLACK_CONNECTOR_SLUG,HumanActionRequiredError,SCAFFOLDABLE_CHANNELS,SLACK_CHANNEL_DEFAULT_ROUTE,SUPPORTED_PROTOCOLS,WriteFileExistsError,byokProviderEnvVar,catalogSlugs,deriveSlackConnectorSlug,effectiveProtocols,endpointForProtocol,ensureChannel,ensureConnection,ensureConnectionDependencies,getCatalogEntry,hasVercelHostFramework,isEveProject,isNextJsProject,isValidConnectionSlug,listAuthoredChannels,listAuthoredConnections,normalizeSlackConnectorSlug,resolveVercelHostFrameworkPreset,scaffoldBaseProject,scaffoldExtensionProject};
+import{WriteFileExistsError}from"./files.js";import{CURRENT_DIRECTORY_PROJECT_NAME,byokProviderEnvVar,isEveProject,scaffoldBaseProject}from"./create/project.js";import{scaffoldExtensionProject}from"./create/extension.js";import{DEFAULT_SLACK_CONNECTOR_SLUG,SLACK_CHANNEL_DEFAULT_ROUTE,deriveSlackConnectorSlug,ensureChannel,hasVercelHostFramework,isNextJsProject,listAuthoredChannels,normalizeSlackConnectorSlug,resolveVercelHostFrameworkPreset}from"./update/channels.js";import{SCAFFOLDABLE_CHANNELS}from"./channels-catalog.js";import{CONNECTION_CATALOG,CUSTOM_CONNECTION_SLUG,SUPPORTED_PROTOCOLS,catalogSlugs,effectiveProtocols,endpointForProtocol,getCatalogEntry,isValidConnectionSlug}from"./connections/catalog.js";import{ensureConnection,ensureConnectionDependencies,listAuthoredConnections}from"./update/connections.js";import{ensurePackageDependencies}from"./update/dependencies.js";import{HumanActionRequiredError}from"../human-action.js";export{CONNECTION_CATALOG,CURRENT_DIRECTORY_PROJECT_NAME,CUSTOM_CONNECTION_SLUG,DEFAULT_SLACK_CONNECTOR_SLUG,HumanActionRequiredError,SCAFFOLDABLE_CHANNELS,SLACK_CHANNEL_DEFAULT_ROUTE,SUPPORTED_PROTOCOLS,WriteFileExistsError,byokProviderEnvVar,catalogSlugs,deriveSlackConnectorSlug,effectiveProtocols,endpointForProtocol,ensureChannel,ensureConnection,ensureConnectionDependencies,ensurePackageDependencies,getCatalogEntry,hasVercelHostFramework,isEveProject,isNextJsProject,isValidConnectionSlug,listAuthoredChannels,listAuthoredConnections,normalizeSlackConnectorSlug,resolveVercelHostFrameworkPreset,scaffoldBaseProject,scaffoldExtensionProject};

package/dist/src/setup/scaffold/update/dependencies.d.ts

@@ -0,0 +1,12 @@
+export interface EnsurePackageDependenciesOptions {
+    readonly dependencies: Readonly<Record<string, string>>;
+    readonly projectRoot: string;
+}
+export interface PackageDependencyMutation {
+    readonly dependencies: readonly string[];
+    readonly devDependencies: readonly string[];
+    readonly path: string;
+    readonly scripts: readonly string[];
+}
+/** Adds missing or outdated runtime dependencies to an existing package.json. */
+export declare function ensurePackageDependencies(options: EnsurePackageDependenciesOptions): Promise<readonly PackageDependencyMutation[]>;

package/dist/src/setup/scaffold/update/dependencies.js

@@ -0,0 +1 @@
+import{patchPackageJson}from"./package-json.js";import{readFile}from"node:fs/promises";import{join}from"node:path";async function ensurePackageDependencies(e){let t=join(e.projectRoot,`package.json`),n;try{n=JSON.parse(await readFile(t,`utf8`))}catch(e){if(e.code===`ENOENT`)return[];throw e}let r=isRecord(n)&&isRecord(n.dependencies)?n.dependencies:{},i=Object.fromEntries(Object.entries(e.dependencies).filter(([e,t])=>r[e]!==t)),a=Object.keys(i);return a.length===0?[]:(await patchPackageJson(t,{dependencies:i}),[{dependencies:a,devDependencies:[],path:t,scripts:[]}])}function isRecord(e){return typeof e==`object`&&!!e&&!Array.isArray(e)}export{ensurePackageDependencies};

package/dist/src/setup/scaffold/version-tokens.d.ts

@@ -1,4 +1,5 @@
 export declare const DEFAULT_CONNECT_PACKAGE_VERSION = "2.2.0";
+export declare const DEFAULT_MICROSANDBOX_PACKAGE_VERSION = "0.5.5";
 /**
  * Returns a scaffold version value, resolving a build-stamp token at the point
  * of need when the running code is unstamped. The published package is stamped

package/dist/src/setup/scaffold/version-tokens.js

@@ -1 +1 @@
-import{dirname,join}from"node:path";import{existsSync,readFileSync,realpathSync}from"node:fs";import{fileURLToPath}from"node:url";function versionToken(e){return`__${e}_VERSION__`}function bareToken(e){return`__${e}__`}const NODE_ENGINE_TOKEN=bareToken(`NODE_ENGINE`),DEFAULT_CONNECT_PACKAGE_VERSION=`2.2.0`,TOKEN_SOURCES={[versionToken(`EVE_PACKAGE`)]:{kind:`eve-version`},[versionToken(`EVE_PACKAGE_DEPENDENCY`)]:{kind:`eve-version`},[NODE_ENGINE_TOKEN]:{kind:`eve-node-engine`},[versionToken(`AI_SDK`)]:{kind:`catalog`,packageName:`ai`},[versionToken(`NEXT`)]:{kind:`catalog`,packageName:`next`},[versionToken(`REACT`)]:{kind:`catalog`,packageName:`react`},[versionToken(`REACT_DOM`)]:{kind:`catalog`,packageName:`react-dom`},[versionToken(`STREAMDOWN`)]:{kind:`catalog`,packageName:`streamdown`},[versionToken(`ZOD`)]:{kind:`catalog`,packageName:`zod`},[versionToken(`TYPESCRIPT`)]:{kind:`catalog`,packageName:`typescript`},[versionToken(`TYPES_REACT`)]:{kind:`catalog`,packageName:`@types/react`},[versionToken(`TYPES_REACT_DOM`)]:{kind:`catalog`,packageName:`@types/react-dom`}},resolvedTokens=new Map;function findEvePackageRoot(){let e=dirname(realpathSync(fileURLToPath(import.meta.url)));for(;;){let t=join(e,`package.json`);if(existsSync(t)&&JSON.parse(readFileSync(t,`utf8`)).name===`eve`)return e;let n=dirname(e);if(n===e)return;e=n}}function findWorkspaceManifest(e){let t=e;for(;;){let e=join(t,`pnpm-workspace.yaml`);if(existsSync(e))return e;let n=dirname(t);if(n===t)return;t=n}}function readCatalogVersion(e,t){let n=readFileSync(e,`utf8`).split(/\r?\n/),r=!1;for(let e of n){if(/^catalog:\s*$/.test(e)){r=!0;continue}if(!r)continue;if(/^\S/.test(e))break;let n=e.match(/^\s+(?:"([^"]+)"|([\w@/.-]+)):\s*"([^"]+)"/);if(n&&(n[1]??n[2])===t)return n[3]}}function resolveTokenFromDevTree(e){let t=TOKEN_SOURCES[e];if(t!==void 0)try{let e=findEvePackageRoot();if(e===void 0||!existsSync(join(e,`scripts/stamp-version-tokens.mjs`)))return;if(t.kind===`eve-version`){let t=JSON.parse(readFileSync(join(e,`package.json`),`utf8`));return typeof t.version==`string`?t.version:void 0}if(t.kind===`eve-node-engine`){let t=JSON.parse(readFileSync(join(e,`package.json`),`utf8`)).engines?.node;return typeof t==`string`?t:void 0}let n=findWorkspaceManifest(e);return n===void 0?void 0:readCatalogVersion(n,t.packageName)}catch{return}}function resolveVersionToken(e,t){if(!t.startsWith(`__`))return t;let n=resolvedTokens.get(t);if(n!==void 0)return n;let r=resolveTokenFromDevTree(t);if(r===void 0)throw Error(`Scaffold received unstamped version token (${e}=${t}). Build eve before using its dist entrypoint.`);return resolvedTokens.set(t,r),r}export{DEFAULT_CONNECT_PACKAGE_VERSION,resolveVersionToken};
+import{dirname,join}from"node:path";import{existsSync,readFileSync,realpathSync}from"node:fs";import{fileURLToPath}from"node:url";function versionToken(e){return`__${e}_VERSION__`}function bareToken(e){return`__${e}__`}const NODE_ENGINE_TOKEN=bareToken(`NODE_ENGINE`),DEFAULT_CONNECT_PACKAGE_VERSION=`2.2.0`,DEFAULT_MICROSANDBOX_PACKAGE_VERSION=`0.5.5`,TOKEN_SOURCES={[versionToken(`EVE_PACKAGE`)]:{kind:`eve-version`},[versionToken(`EVE_PACKAGE_DEPENDENCY`)]:{kind:`eve-version`},[NODE_ENGINE_TOKEN]:{kind:`eve-node-engine`},[versionToken(`AI_SDK`)]:{kind:`catalog`,packageName:`ai`},[versionToken(`NEXT`)]:{kind:`catalog`,packageName:`next`},[versionToken(`REACT`)]:{kind:`catalog`,packageName:`react`},[versionToken(`REACT_DOM`)]:{kind:`catalog`,packageName:`react-dom`},[versionToken(`STREAMDOWN`)]:{kind:`catalog`,packageName:`streamdown`},[versionToken(`ZOD`)]:{kind:`catalog`,packageName:`zod`},[versionToken(`TYPESCRIPT`)]:{kind:`catalog`,packageName:`typescript`},[versionToken(`TYPES_REACT`)]:{kind:`catalog`,packageName:`@types/react`},[versionToken(`TYPES_REACT_DOM`)]:{kind:`catalog`,packageName:`@types/react-dom`}},resolvedTokens=new Map;function findEvePackageRoot(){let e=dirname(realpathSync(fileURLToPath(import.meta.url)));for(;;){let t=join(e,`package.json`);if(existsSync(t)&&JSON.parse(readFileSync(t,`utf8`)).name===`eve`)return e;let n=dirname(e);if(n===e)return;e=n}}function findWorkspaceManifest(e){let t=e;for(;;){let e=join(t,`pnpm-workspace.yaml`);if(existsSync(e))return e;let n=dirname(t);if(n===t)return;t=n}}function readCatalogVersion(e,t){let n=readFileSync(e,`utf8`).split(/\r?\n/),r=!1;for(let e of n){if(/^catalog:\s*$/.test(e)){r=!0;continue}if(!r)continue;if(/^\S/.test(e))break;let n=e.match(/^\s+(?:"([^"]+)"|([\w@/.-]+)):\s*"([^"]+)"/);if(n&&(n[1]??n[2])===t)return n[3]}}function resolveTokenFromDevTree(e){let t=TOKEN_SOURCES[e];if(t!==void 0)try{let e=findEvePackageRoot();if(e===void 0||!existsSync(join(e,`scripts/stamp-version-tokens.mjs`)))return;if(t.kind===`eve-version`){let t=JSON.parse(readFileSync(join(e,`package.json`),`utf8`));return typeof t.version==`string`?t.version:void 0}if(t.kind===`eve-node-engine`){let t=JSON.parse(readFileSync(join(e,`package.json`),`utf8`)).engines?.node;return typeof t==`string`?t:void 0}let n=findWorkspaceManifest(e);return n===void 0?void 0:readCatalogVersion(n,t.packageName)}catch{return}}function resolveVersionToken(e,t){if(!t.startsWith(`__`))return t;let n=resolvedTokens.get(t);if(n!==void 0)return n;let r=resolveTokenFromDevTree(t);if(r===void 0)throw Error(`Scaffold received unstamped version token (${e}=${t}). Build eve before using its dist entrypoint.`);return resolvedTokens.set(t,r),r}export{DEFAULT_CONNECT_PACKAGE_VERSION,DEFAULT_MICROSANDBOX_PACKAGE_VERSION,resolveVersionToken};

package/dist/src/shared/agent-definition.d.ts

@@ -180,15 +180,6 @@ export interface AgentLimitsDefinition {
  * These options are unstable and may change or be removed in any release.
  */
 export interface AgentExperimentalDefinition {
-    /**
-     * Reads instrumentation from an `instrumentation/` directory of providers
-     * rather than a single `agent/instrumentation.ts` config object.
-     *
-     * The two layouts are mutually exclusive: with this on, an
-     * `agent/instrumentation.ts` is a build error, and with it off, an
-     * `instrumentation/` directory is.
-     */
-    readonly instrumentationProviders?: boolean;
     /**
      * Runs this agent's delegated subagent calls as durable background tasks.
      * The originating tool call returns a task receipt immediately and the

package/dist/src/shared/trace-policy.d.ts

@@ -1,7 +1,5 @@
 import type { ConversationContext } from "#shared/conversation-context.js";
 import { type InstrumentationDecision } from "#shared/instrumentation-decision.js";
-/** @deprecated Use `TraceCapturePolicy` to select directional content. */
-export type InstrumentationCapture = "content" | "metadata";
 export type TraceCaptureContext = {
     readonly agentName: string;
 } & ConversationContext;
@@ -13,7 +11,6 @@ export type TracePolicyDecision = {
     readonly recordOutputs: boolean;
 };
 export type TraceCapturePolicy = (trace: TraceCaptureContext) => TracePolicyDecision | boolean;
-export declare function legacyCaptureTracePolicy(capture: InstrumentationCapture | undefined): TraceCapturePolicy | undefined;
 export declare function resolveTracePolicy(policy: TraceCapturePolicy | undefined, trace: TraceCaptureContext, onError?: (error: unknown) => void): InstrumentationDecision;
 export declare function resolveTracePolicyDecision(decision: TracePolicyDecision | boolean, context: {
     readonly audience: TraceCaptureContext["audience"];

package/dist/src/shared/trace-policy.js

@@ -1 +1 @@
-import{shouldCaptureInstrumentationContent}from"#shared/instrumentation-content.js";import{DROP_INSTRUMENTATION}from"#shared/instrumentation-decision.js";function legacyCaptureTracePolicy(e){if(e===void 0)return;let t=e===`content`;return()=>({emit:!0,recordInputs:t,recordOutputs:t})}function resolveTracePolicy(e,t,n){try{let n=e?.(t);return resolveTracePolicyDecision(n??{emit:!0,recordInputs:t.audience===`public`||t.environment===`development`,recordOutputs:t.audience===`public`||t.environment===`development`},t)}catch(e){try{n?.(e)}catch{}return DROP_INSTRUMENTATION}}function resolveTracePolicyDecision(e,t){if(e===!1)return DROP_INSTRUMENTATION;if(e===!0){let e=shouldCaptureInstrumentationContent(t);return{action:`record`,recordInputs:e,recordOutputs:e}}return e.emit?{action:`record`,recordInputs:e.recordInputs,recordOutputs:e.recordOutputs}:DROP_INSTRUMENTATION}export{legacyCaptureTracePolicy,resolveTracePolicy,resolveTracePolicyDecision};
+import{shouldCaptureInstrumentationContent}from"#shared/instrumentation-content.js";import{DROP_INSTRUMENTATION}from"#shared/instrumentation-decision.js";function resolveTracePolicy(e,t,n){try{let n=e?.(t);return resolveTracePolicyDecision(n??{emit:!0,recordInputs:t.audience===`public`||t.environment===`development`,recordOutputs:t.audience===`public`||t.environment===`development`},t)}catch(e){try{n?.(e)}catch{}return DROP_INSTRUMENTATION}}function resolveTracePolicyDecision(e,t){if(e===!1)return DROP_INSTRUMENTATION;if(e===!0){let e=shouldCaptureInstrumentationContent(t);return{action:`record`,recordInputs:e,recordOutputs:e}}return e.emit?{action:`record`,recordInputs:e.recordInputs,recordOutputs:e.recordOutputs}:DROP_INSTRUMENTATION}export{resolveTracePolicy,resolveTracePolicyDecision};

package/dist/src/tools/workflow-definition.d.ts

@@ -42,6 +42,8 @@ type JsonSchemaOutput<TSchema> = TSchema extends {
 export interface WorkflowAgentMetadata {
     readonly description: string;
 }
+/** Context capabilities available inside an authored `"use step"` helper. */
+export type WorkflowStepToolContext = Pick<ToolContext, "abortSignal" | "callId" | "session" | "toolName" | "getToken" | "requireAuth">;
 interface WorkflowAgent {
     <const TOutputSchema extends JsonObject>(target: string, input: AgentInput & {
         readonly outputSchema: TOutputSchema;
@@ -49,8 +51,8 @@ interface WorkflowAgent {
     (target: string, input: AgentInput): Promise<JsonValue>;
 }
 /**
- * Context supplied to a workflow tool. Pass it directly to a step helper for
- * getToken/requireAuth; those capabilities throw in the workflow body itself.
+ * Context supplied to a workflow tool body. When passed directly to a step,
+ * eve replaces it with {@link WorkflowStepToolContext}.
  */
 export type WorkflowToolContext = Pick<ToolContext, "abortSignal" | "callId" | "session" | "toolName" | "getToken" | "requireAuth"> & {
     /** Invoke an agent by its invocation name. */

package/dist/src/tracing/content-span-processor.js

@@ -1 +1 @@
-import{normalizeChannelAudience}from"#shared/channel-audience.js";import{hasConversationRelease}from"#tracing/local-traces.js";import{channelAudienceFromContext}from"#tracing/channel-audience-context.js";import{withoutDeclinedContent}from"#tracing/content-attributes.js";import{normalizeSpanExportPolicies}from"#tracing/span-export-policy.js";function contentFilteringProcessor(e,t){let n=normalizeSpanExportPolicies(t);if(n.length===0)return e;let r=e;for(let e of n.toReversed())r=policyFilteringProcessor(r,e);return r}function policyFilteringProcessor(e,t){let n=new WeakMap,r=new WeakSet,i={forceFlush:()=>e.forceFlush(),onEnd:i=>{if(typeof i!=`object`||!i)return;let a=facadeFor(i,t,n);if(r.has(i)||!a.exported){r.delete(i),n.delete(i);return}a.refresh();try{e.onEnd(a.value)}finally{n.delete(i)}},onStart:(i,a)=>{if(typeof i!=`object`||!i)return;let o=facadeFor(i,t,n,channelAudienceFromContext(a));if(!o.exported){r.add(i);return}e.onStart(o.value,a)},shutdown:()=>e.shutdown()};return hasConversationRelease(e)?{...i,releaseConversation:t=>e.releaseConversation(t)}:i}function facadeFor(e,t,n,r){let i=n.get(e);if(i!==void 0)return i;let a=spanExportContext(e,r),o=spanExportDecision(a,t),s={recordInputs:!o.redactInputs,recordOutputs:!o.redactOutputs},c={},l=[],u={},d=Object.create(Reflect.getPrototypeOf(e)),f=new Map,refresh=()=>{refreshAttributes(c,e,s,a,t),refreshEvents(l,e,s),refreshStatus(u,e,s)},p,readOriginal=t=>{let n=Reflect.get(e,t,e);if(typeof n!=`function`||t===`constructor`)return n;let r=f.get(t);if(r!==void 0)return r;let created=(...t)=>{let r=Reflect.apply(n,e,t);return refresh(),r===e?p:r};return f.set(t,created),created};Object.defineProperty(d,"attributes",{configurable:!0,enumerable:!0,value:c}),Object.defineProperty(d,"events",{configurable:!0,enumerable:!0,value:l}),Object.defineProperty(d,"status",{configurable:!0,enumerable:!0,value:u});for(let t of Reflect.ownKeys(e)){if(t===`attributes`||t===`events`||t===`status`)continue;let n=Reflect.getOwnPropertyDescriptor(e,t);Object.defineProperty(d,t,{configurable:!0,enumerable:n?.enumerable??!1,get:()=>readOriginal(t)})}p=new Proxy(d,{get:(e,t,n)=>Object.hasOwn(e,t)?Reflect.get(e,t,n):readOriginal(t)});let m={context:a,exported:o.exported,refresh,value:p};return m.refresh(),n.set(e,m),m}function refreshAttributes(e,t,n,r,i){for(let t of Object.keys(e))delete e[t];let a=t.attributes;if(typeof a!=`object`||!a)return;let o=withoutDeclinedContent(a,n)??a;for(let[t,n]of Object.entries(o)){let a=attributeDecision(i,{key:t,span:r,value:n});typeof a==`object`&&a&&(`emit`in a&&`replace`in a||(`replace`in a&&a.replace===!0?a.value!==void 0&&(e[t]=a.value):`emit`in a&&a.emit===!0&&(e[t]=n)))}}function spanExportContext(e,t=`unknown`){let n=e.attributes,r=typeof n==`object`&&n?n:{},i=[r[`agent.channel.audience`],r[`ai.settings.context.eve.channel.audience`]].filter(e=>e!==void 0).map(normalizeChannelAudience),a=i.length>0&&i.every(e=>e===i[0])?i[0]:i.length>0?`unknown`:t,o=e.spanContext?.(),s=typeof o==`object`&&o?o:{},c=e.name;return{attributes:r,audience:a,name:typeof c==`string`?c:``,spanId:typeof s.spanId==`string`?s.spanId:``,traceId:typeof s.traceId==`string`?s.traceId:``}}function spanExportDecision(e,t){let n;try{n=t.span?.(e)??{emit:!0}}catch{return{exported:!1,redactInputs:!1,redactOutputs:!1}}if(typeof n==`boolean`)return{exported:n,redactInputs:!1,redactOutputs:!1};if(typeof n!=`object`||!n)return{exported:!1,redactInputs:!1,redactOutputs:!1};if(`emit`in n)return`redact`in n?{exported:!1,redactInputs:!1,redactOutputs:!1}:{exported:typeof n.emit==`boolean`&&n.emit,redactInputs:!1,redactOutputs:!1};if(n.redact!==!0)return{exported:!1,redactInputs:!1,redactOutputs:!1};let r=n.inputs===!0,i=n.outputs===!0;return!r&&!i?{exported:!1,redactInputs:!1,redactOutputs:!1}:{exported:!0,redactInputs:r,redactOutputs:i}}function attributeDecision(e,t){if(e?.attribute===void 0)return{emit:!0};try{return e.attribute(t)}catch{return{emit:!1}}}function refreshEvents(e,t,n){e.length=0;let r=t.events;if(Array.isArray(r)){if(n.recordOutputs){e.push(...r);return}for(let t of r){if(typeof t!=`object`||!t)continue;let n=t;n.name!==`exception`&&e.push({...n,attributes:void 0})}}}function refreshStatus(e,t,n){for(let t of Object.keys(e))delete e[t];let r=t.status;if(typeof r!=`object`||!r)return;let i=r;i.code!==void 0&&(e.code=i.code),n.recordOutputs&&i.message!==void 0&&(e.message=i.message)}export{contentFilteringProcessor};
+import{normalizeChannelAudience}from"#shared/channel-audience.js";import{hasConversationRelease}from"#tracing/local-traces.js";import{channelAudienceFromContext}from"#tracing/channel-audience-context.js";import{withoutDeclinedContent}from"#tracing/content-attributes.js";import{normalizeSpanExportPolicies}from"#tracing/span-export-policy.js";function contentFilteringProcessor(e,t){let n=normalizeSpanExportPolicies(t);if(n.length===0)return e;let r=e;for(let e of n.toReversed())r=policyFilteringProcessor(r,e);return r}function policyFilteringProcessor(e,t){let n=new WeakMap,r=new WeakSet,i={forceFlush:()=>e.forceFlush(),onEnd:i=>{if(typeof i!=`object`||!i)return;let a=facadeFor(i,t,n);if(r.has(i)||!a.exported){r.delete(i),n.delete(i);return}a.refresh();try{e.onEnd(a.value)}finally{n.delete(i)}},onStart:(i,a)=>{if(typeof i!=`object`||!i)return;let o=facadeFor(i,t,n,channelAudienceFromContext(a));if(!o.exported){r.add(i);return}e.onStart(o.value,a)},shutdown:()=>e.shutdown()};return hasConversationRelease(e)?{...i,releaseConversation:t=>e.releaseConversation(t)}:i}function facadeFor(e,t,n,r){let i=n.get(e);if(i!==void 0)return i;let a=spanExportContext(e,r),o=spanExportDecision(a,t),s={recordInputs:!o.redactInputs,recordOutputs:!o.redactOutputs},c={},l=[],u={},d=Object.create(Reflect.getPrototypeOf(e)),f=new Map,refresh=()=>{refreshAttributes(c,e,s,a,t),refreshEvents(l,e,s),refreshStatus(u,e,s)},p,readOriginal=t=>{let n=Reflect.get(e,t,e);if(typeof n!=`function`||t===`constructor`)return n;let r=f.get(t);if(r!==void 0)return r;let created=(...t)=>{let r=Reflect.apply(n,e,t);return refresh(),r===e?p:r};return f.set(t,created),created};Object.defineProperty(d,"attributes",{configurable:!0,enumerable:!0,value:c}),Object.defineProperty(d,"events",{configurable:!0,enumerable:!0,value:l}),Object.defineProperty(d,"status",{configurable:!0,enumerable:!0,value:u});for(let t of Reflect.ownKeys(e)){if(t===`attributes`||t===`events`||t===`status`)continue;let n=Reflect.getOwnPropertyDescriptor(e,t);Object.defineProperty(d,t,{configurable:!0,enumerable:n?.enumerable??!1,get:()=>readOriginal(t)})}p=new Proxy(d,{get:(e,t,n)=>Object.hasOwn(e,t)?Reflect.get(e,t,n):readOriginal(t)});let m={context:a,exported:o.exported,refresh,value:p};return m.refresh(),n.set(e,m),m}function refreshAttributes(e,t,n,r,i){for(let t of Object.keys(e))delete e[t];let a=t.attributes;if(typeof a!=`object`||!a)return;let o=withoutDeclinedContent(a,n)??a;for(let[t,n]of Object.entries(o)){let a=attributeDecision(i,{key:t,span:r,value:n});typeof a==`object`&&a&&(`emit`in a&&`replace`in a||(`replace`in a&&a.replace===!0?a.value!==void 0&&(e[t]=a.value):`emit`in a&&a.emit===!0&&(e[t]=n)))}}function spanExportContext(e,t=`unknown`){let n=e.attributes,r=typeof n==`object`&&n?n:{},i=[r[`agent.channel.audience`],r[`ai.settings.context.eve.channel.audience`]].filter(e=>e!==void 0).map(normalizeChannelAudience),a=i.length>0&&i.every(e=>e===i[0])?i[0]:i.length>0?`unknown`:t,o=e.spanContext?.(),s=typeof o==`object`&&o?o:{},c=e.name;return{attributes:r,audience:a,name:typeof c==`string`?c:``,spanId:typeof s.spanId==`string`?s.spanId:``,traceId:typeof s.traceId==`string`?s.traceId:``}}function spanExportDecision(e,t){let n;try{n=t.span?.(e)??{emit:!0}}catch{return{exported:!1,redactInputs:!1,redactOutputs:!1}}if(typeof n!=`object`||!n)return{exported:!1,redactInputs:!1,redactOutputs:!1};if(`emit`in n)return`redact`in n?{exported:!1,redactInputs:!1,redactOutputs:!1}:{exported:typeof n.emit==`boolean`&&n.emit,redactInputs:!1,redactOutputs:!1};if(n.redact!==!0)return{exported:!1,redactInputs:!1,redactOutputs:!1};let r=n.inputs===!0,i=n.outputs===!0;return!r&&!i?{exported:!1,redactInputs:!1,redactOutputs:!1}:{exported:!0,redactInputs:r,redactOutputs:i}}function attributeDecision(e,t){if(e?.attribute===void 0)return{emit:!0};try{return e.attribute(t)}catch{return{emit:!1}}}function refreshEvents(e,t,n){e.length=0;let r=t.events;if(Array.isArray(r)){if(n.recordOutputs){e.push(...r);return}for(let t of r){if(typeof t!=`object`||!t)continue;let n=t;n.name!==`exception`&&e.push({...n,attributes:void 0})}}}function refreshStatus(e,t,n){for(let t of Object.keys(e))delete e[t];let r=t.status;if(typeof r!=`object`||!r)return;let i=r;i.code!==void 0&&(e.code=i.code),n.recordOutputs&&i.message!==void 0&&(e.message=i.message)}export{contentFilteringProcessor};

package/dist/src/tracing/install-instrumentation-runtime.d.ts

@@ -4,9 +4,8 @@ import type { CollectedOtel, RuntimeContextResolver } from "#tracing/otel-declar
 /**
  * Installs the process instrumentation runtime around a collected pipeline.
  *
- * Both layouts land here. `eve dev`'s zero-config default and an authored
- * `agent/instrumentation/` directory differ only in where the declared values
- * came from, so sharing the install keeps them on one runtime path.
+ * `eve dev`'s zero-config default and an authored `agent/instrumentation/`
+ * directory differ only in where the declared values came from.
  *
  * A directory that declared no OpenTelemetry still gets a bus: its providers
  * see every event, they just have no spans to hang them on.
@@ -14,7 +13,6 @@ import type { CollectedOtel, RuntimeContextResolver } from "#tracing/otel-declar
 export declare function installInstrumentationRuntime(input: {
     readonly collected: CollectedOtel;
     readonly frameworkVersion: string;
-    readonly instrumentationProviders?: boolean;
     readonly providers: readonly InstrumentationProviderDefinition[];
     readonly runtimeContextResolvers?: readonly RuntimeContextResolver[];
     readonly serviceName: string;

package/dist/src/tracing/install-instrumentation-runtime.js

@@ -1 +1 @@
-import{createLogger,formatError}from"#internal/logging.js";import{contextStorage}from"#context/container.js";import{ConversationIdKey}from"#context/keys.js";import{resolveInstrumentationEnvironment}from"#internal/application/dev-environment.js";import{readConversationId}from"#tracing/conversation-context.js";import{ContextAgentTraceStateStore}from"#tracing/agent-trace-context-store.js";import{registerInstrumentationRuntime}from"#instrumentation/runtime.js";import{trace}from"#compiled/@opentelemetry/api/index.js";import{createInstrumentationHooks}from"#instrumentation/lifecycle.js";import{AgentSpanIdGenerator}from"#tracing/agent-span-id-generator.js";import{hasConversationRelease}from"#tracing/local-traces.js";import{createAgentOtelInstrumentation}from"#tracing/agent-otel-provider.js";import{registerOtelPipeline}from"#tracing/otel-registration.js";const log=createLogger(`tracing.install-instrumentation-runtime`);function installInstrumentationRuntime(e){let t=[],n=[],r,flushSettledInvocations,i,a,runInContext=(e,t)=>t();if(e.collected.declared){r=registerOtelPipeline({pipeline:e.collected.pipeline,serviceName:e.serviceName});let o=createAgentOtelInstrumentation({environment:resolveInstrumentationEnvironment(),frameworkVersion:e.frameworkVersion,idGenerator:r.idGenerator,recordInputs:e.collected.settings.recordInputs,recordOutputs:e.collected.settings.recordOutputs,samplesTrace:r.samplesTrace,stateStore:new ContextAgentTraceStateStore,tracer:trace.getTracer(`eve.agent`,e.frameworkVersion),tracePolicy:e.collected.settings.tracePolicy});t.push({...o.hook,stateNamespace:`internal:otel`}),flushSettledInvocations=async()=>{await o.hook.flush?.()},i=o.prepareSessionTrace,a=o.prepareTurnTrace,runInContext=o.runInContext;let s=e.collected.pipeline.spanProcessors.filter(isSpanProcessor).filter(hasConversationRelease);s.length>0&&n.push(sessionReleaseProvider(s))}let o=[...t,...e.providers,...n],s;return registerInstrumentationRuntime({flushSettledInvocations,forceFlush:async()=>{await settleAll(o.map(e=>()=>e.flush?.())),await settleAll(r===void 0?[]:[r.forceFlush])},hooks:createInstrumentationHooks({parallel:e.providers,serialAfter:n,serialBefore:t}),idGenerator:r?.idGenerator??new AgentSpanIdGenerator,instrumentationProviders:e.instrumentationProviders,memoryOperations:r!==void 0||e.providers.some(hasMemoryOperationHandler),otelSettings:e.collected.declared?e.collected.settings:void 0,ownsAgentSpans:r!==void 0,prepareSessionTrace:i,prepareTurnTrace:a,runtimeContextResolvers:e.runtimeContextResolvers,runInContext,samplesTrace:r?.samplesTrace,shutdown:()=>(s??=(async()=>{await settleAll(o.map(e=>()=>e.shutdown?.())),await settleAll(r===void 0?[]:[r.shutdown])})(),s)})}function hasMemoryOperationHandler(e){return e.events?.[`memory.operation.started`]!==void 0||e.events?.[`memory.operation.completed`]!==void 0||e.events?.[`memory.operation.failed`]!==void 0}function isSpanProcessor(e){return e!==`auto`}function sessionReleaseProvider(e){let release=async t=>{let n=readConversationId(contextStorage.getStore()?.get(ConversationIdKey))??t.sessionId;n===t.sessionId&&await Promise.all(e.map(e=>e.releaseConversation(n)))};return{events:{"session.completed":release,"session.failed":release},name:`eve.conversation-release`,stateNamespace:`internal:conversation-release`}}async function settleAll(e){let t=await Promise.allSettled(e.map(async e=>e()));for(let e of t)e.status===`rejected`&&log.warn(`instrumentation drain failed`,{error:formatError(e.reason)})}export{installInstrumentationRuntime};
+import{createLogger,formatError}from"#internal/logging.js";import{contextStorage}from"#context/container.js";import{ConversationIdKey}from"#context/keys.js";import{resolveInstrumentationEnvironment}from"#internal/application/dev-environment.js";import{readConversationId}from"#tracing/conversation-context.js";import{ContextAgentTraceStateStore}from"#tracing/agent-trace-context-store.js";import{registerInstrumentationRuntime}from"#instrumentation/runtime.js";import{trace}from"#compiled/@opentelemetry/api/index.js";import{createInstrumentationHooks}from"#instrumentation/lifecycle.js";import{AgentSpanIdGenerator}from"#tracing/agent-span-id-generator.js";import{hasConversationRelease}from"#tracing/local-traces.js";import{createAgentOtelInstrumentation}from"#tracing/agent-otel-provider.js";import{registerOtelPipeline}from"#tracing/otel-registration.js";const log=createLogger(`tracing.install-instrumentation-runtime`);function installInstrumentationRuntime(e){let t=[],n=[],r,flushSettledInvocations,i,a,runInContext=(e,t)=>t();if(e.collected.declared){r=registerOtelPipeline({pipeline:e.collected.pipeline,serviceName:e.serviceName});let o=createAgentOtelInstrumentation({environment:resolveInstrumentationEnvironment(),frameworkVersion:e.frameworkVersion,idGenerator:r.idGenerator,recordInputs:e.collected.settings.recordInputs,recordOutputs:e.collected.settings.recordOutputs,samplesTrace:r.samplesTrace,stateStore:new ContextAgentTraceStateStore,tracer:trace.getTracer(`eve.agent`,e.frameworkVersion),tracePolicy:e.collected.settings.tracePolicy});t.push({...o.hook,stateNamespace:`internal:otel`}),flushSettledInvocations=async()=>{await o.hook.flush?.()},i=o.prepareSessionTrace,a=o.prepareTurnTrace,runInContext=o.runInContext;let s=e.collected.pipeline.spanProcessors.filter(isSpanProcessor).filter(hasConversationRelease);s.length>0&&n.push(sessionReleaseProvider(s))}let o=[...t,...e.providers,...n],s;return registerInstrumentationRuntime({flushSettledInvocations,forceFlush:async()=>{await settleAll(o.map(e=>()=>e.flush?.())),await settleAll(r===void 0?[]:[r.forceFlush])},hooks:createInstrumentationHooks({parallel:e.providers,serialAfter:n,serialBefore:t}),idGenerator:r?.idGenerator??new AgentSpanIdGenerator,memoryOperations:r!==void 0||e.providers.some(hasMemoryOperationHandler),otelSettings:e.collected.declared?e.collected.settings:void 0,ownsAgentSpans:r!==void 0,prepareSessionTrace:i,prepareTurnTrace:a,runtimeContextResolvers:e.runtimeContextResolvers,runInContext,samplesTrace:r?.samplesTrace,shutdown:()=>(s??=(async()=>{await settleAll(o.map(e=>()=>e.shutdown?.())),await settleAll(r===void 0?[]:[r.shutdown])})(),s)})}function hasMemoryOperationHandler(e){return e.events?.[`memory.operation.started`]!==void 0||e.events?.[`memory.operation.completed`]!==void 0||e.events?.[`memory.operation.failed`]!==void 0}function isSpanProcessor(e){return e!==`auto`}function sessionReleaseProvider(e){let release=async t=>{let n=readConversationId(contextStorage.getStore()?.get(ConversationIdKey))??t.sessionId;n===t.sessionId&&await Promise.all(e.map(e=>e.releaseConversation(n)))};return{events:{"session.completed":release,"session.failed":release},name:`eve.conversation-release`,stateNamespace:`internal:conversation-release`}}async function settleAll(e){let t=await Promise.allSettled(e.map(async e=>e()));for(let e of t)e.status===`rejected`&&log.warn(`instrumentation drain failed`,{error:formatError(e.reason)})}export{installInstrumentationRuntime};

package/dist/src/tracing/otel-declaration.d.ts

@@ -42,7 +42,7 @@ export interface OtelOptions {
      * settings can only narrow it. For a trusted remote trace, eve evaluates this
      * policy against the immutable origin audience and intersects it with the
      * parent's effective ceiling, so every hop can only narrow capture. It never
-     * changes what authored instrumentation providers receive. A thrown error rejects trace production without
+     * changes what lifecycle instrumentation receives. A thrown error rejects trace production without
      * silencing lifecycle providers. The policy may be invoked more than once
      * while a new session is being prepared, so it must be deterministic for
      * consistent results.
@@ -77,7 +77,7 @@ export interface ManagedTraceOptions {
     readonly exportPolicy?: SpanExportPolicy | readonly SpanExportPolicy[];
 }
 /** Where one `otelIntegration()` sends spans and metrics. */
-export interface OtelIntegrationOptions {
+export interface OtelIntegrationOptions extends ManagedTraceOptions {
     /** Merged into the pipeline in declaration order. */
     readonly spanProcessors?: readonly SpanProcessor[];
     /** Wrapped in eve's batching processor and appended after `spanProcessors`. */
@@ -138,7 +138,7 @@ export declare function otel(options?: OtelOptions): OtelDeclaration;
  */
 export declare function otelIntegration(options?: OtelIntegrationOptions): OtelIntegration;
 /** @internal Local and Agent Runs destination declaration. */
-export declare function managedOtelIntegration(options?: OtelIntegrationOptions & ManagedTraceOptions): OtelIntegration;
+export declare function managedOtelIntegration(options?: OtelIntegrationOptions): OtelIntegration;
 /** Vercel Agent Runs through the hosted request-context transport. @internal */
 export declare function agentRunsIntegration(options?: ManagedTraceOptions): OtelIntegration;
 export declare function isOtelDeclaration(value: unknown): value is OtelDeclaration;
@@ -157,8 +157,9 @@ export interface OtelHarnessSettings {
     readonly functionId?: string;
     readonly traceChannelRequests: boolean;
     readonly tracePolicy?: TraceCapturePolicy;
-    /** Legacy `defineInstrumentation()` capture settings. Provider destinations capture fully. */
+    /** Whether AI SDK spans include model and tool inputs. */
     readonly recordInputs: boolean;
+    /** Whether AI SDK spans include model and tool outputs. */
     readonly recordOutputs: boolean;
 }
 /** @internal */

package/dist/src/tracing/otel-declaration.js

@@ -1 +1 @@
-import{PROVIDER}from"#public/instrumentation/provider.js";import{batchSpanProcessor}from"#tracing/batch-span-processor.js";import{contentFilteringProcessor}from"#tracing/content-span-processor.js";import{vercelRuntimeSpanProcessor}from"#tracing/vercel-runtime-span-exporter.js";const OTEL_DECLARATION=Symbol.for(`eve.instrumentation.otel`),OTEL_INTEGRATION=Symbol.for(`eve.instrumentation.otel-integration`);function otel(e={}){return{[OTEL_DECLARATION]:!0,[PROVIDER]:!0,options:e}}function otelIntegration(e={}){return createOtelIntegration(e)}function managedOtelIntegration(e={}){return createOtelIntegration(e,e.exportPolicy)}function createOtelIntegration(e,t){assertNoRemovedContentOptions(e);let n=e.spanProcessors??[],r=e.traceExporter===void 0?n:[...n,batchSpanProcessor(e.traceExporter)];return{[OTEL_INTEGRATION]:!0,[PROVIDER]:!0,metricReaders:e.metricReaders??[],runtimeContext:e.runtimeContext,spanProcessors:r.map(e=>contentFilteringProcessor(e,t))}}function agentRunsIntegration(e={}){return assertNoRemovedContentOptions(e),{[OTEL_INTEGRATION]:!0,[PROVIDER]:!0,metricReaders:[],spanProcessors:[contentFilteringProcessor(vercelRuntimeSpanProcessor(),e.exportPolicy)]}}function assertNoRemovedContentOptions(e){if(Object.hasOwn(e,`recordInputs`)||Object.hasOwn(e,`recordOutputs`))throw Error("OpenTelemetry destination options no longer support `recordInputs` or `recordOutputs`. Use an `exportPolicy` span decision with `{ redact: true, inputs: true }`, `{ redact: true, outputs: true }`, or both.")}function isOtelDeclaration(e){return typeof e==`object`&&!!e&&e[OTEL_DECLARATION]===!0}function isOtelIntegration(e){return typeof e==`object`&&!!e&&e[OTEL_INTEGRATION]===!0}function collectOtelPipeline(e){let t=[],n=[],r=[],i,a=!1,o=!1;for(let s of e){if(isOtelIntegration(s)){a=!0,o=!0,t.push(...s.spanProcessors),n.push(...s.metricReaders),s.runtimeContext!==void 0&&r.push(s.runtimeContext);continue}if(isOtelDeclaration(s)){if(i!==void 0)throw Error("Instrumentation declares `otel()` more than once. One process has one OpenTelemetry tracer provider, so it has one resource, one sampler, and one propagator set — declare them in a single `otel()`.");a=!0,i=s}}let s=i?.options??{},c={functionId:s.functionId,recordInputs:o,recordOutputs:o,traceChannelRequests:s.traceChannelRequests===!0};return s.tracePolicy!==void 0&&Object.assign(c,{tracePolicy:s.tracePolicy}),{declared:a,pipeline:{instrumentations:s.instrumentations,metricReaders:n.length>0?n:void 0,propagators:s.propagators,resource:s.resource,sampler:s.sampler,spanProcessors:t},runtimeContextResolvers:r,settings:c}}export{agentRunsIntegration,collectOtelPipeline,isOtelDeclaration,isOtelIntegration,managedOtelIntegration,otel,otelIntegration};
+import{PROVIDER}from"#public/instrumentation/provider.js";import{batchSpanProcessor}from"#tracing/batch-span-processor.js";import{contentFilteringProcessor}from"#tracing/content-span-processor.js";import{vercelRuntimeSpanProcessor}from"#tracing/vercel-runtime-span-exporter.js";const OTEL_DECLARATION=Symbol.for(`eve.instrumentation.otel`),OTEL_INTEGRATION=Symbol.for(`eve.instrumentation.otel-integration`);function otel(e={}){return{[OTEL_DECLARATION]:!0,[PROVIDER]:!0,options:e}}function otelIntegration(e={}){return createOtelIntegration(e)}function managedOtelIntegration(e={}){return createOtelIntegration(e)}function createOtelIntegration(e){assertNoRemovedContentOptions(e);let t=e.spanProcessors??[],n=e.traceExporter===void 0?t:[...t,batchSpanProcessor(e.traceExporter)];return{[OTEL_INTEGRATION]:!0,[PROVIDER]:!0,metricReaders:e.metricReaders??[],runtimeContext:e.runtimeContext,spanProcessors:n.map(t=>contentFilteringProcessor(t,e.exportPolicy))}}function agentRunsIntegration(e={}){return assertNoRemovedContentOptions(e),{[OTEL_INTEGRATION]:!0,[PROVIDER]:!0,metricReaders:[],spanProcessors:[contentFilteringProcessor(vercelRuntimeSpanProcessor(),e.exportPolicy)]}}function assertNoRemovedContentOptions(e){if(Object.hasOwn(e,`recordInputs`)||Object.hasOwn(e,`recordOutputs`))throw Error("OpenTelemetry destination options no longer support `recordInputs` or `recordOutputs`. Use an `exportPolicy` span decision with `{ redact: true, inputs: true }`, `{ redact: true, outputs: true }`, or both.")}function isOtelDeclaration(e){return typeof e==`object`&&!!e&&e[OTEL_DECLARATION]===!0}function isOtelIntegration(e){return typeof e==`object`&&!!e&&e[OTEL_INTEGRATION]===!0}function collectOtelPipeline(e){let t=[],n=[],r=[],i,a=!1,o=!1;for(let s of e){if(isOtelIntegration(s)){a=!0,o=!0,t.push(...s.spanProcessors),n.push(...s.metricReaders),s.runtimeContext!==void 0&&r.push(s.runtimeContext);continue}if(isOtelDeclaration(s)){if(i!==void 0)throw Error("Instrumentation declares `otel()` more than once. One process has one OpenTelemetry tracer provider, so it has one resource, one sampler, and one propagator set — declare them in a single `otel()`.");a=!0,i=s}}let s=i?.options??{},c={functionId:s.functionId,recordInputs:o,recordOutputs:o,traceChannelRequests:s.traceChannelRequests===!0};return s.tracePolicy!==void 0&&Object.assign(c,{tracePolicy:s.tracePolicy}),{declared:a,pipeline:{instrumentations:s.instrumentations,metricReaders:n.length>0?n:void 0,propagators:s.propagators,resource:s.resource,sampler:s.sampler,spanProcessors:t},runtimeContextResolvers:r,settings:c}}export{agentRunsIntegration,collectOtelPipeline,isOtelDeclaration,isOtelIntegration,managedOtelIntegration,otel,otelIntegration};

package/dist/src/tracing/span-export-policy.d.ts

@@ -6,9 +6,7 @@ export interface SpanExportContext {
     readonly spanId: string;
     readonly traceId: string;
 }
-export type SpanExportDecision = 
-/** @deprecated Return `{ emit: boolean }` instead. */
-boolean | {
+export type SpanExportDecision = {
     readonly emit: boolean;
 } | {
     readonly redact: true;

package/docs/agent-config.md

@@ -350,7 +350,7 @@ During `eve dev`, ordinary dependencies are bundled into each retained runtime g
 | Per-tool approval (HITL)      | `agent/tools/*.ts`, [Tools](./tools)                                             |
 | Inbound auth & network policy | the channel layer, [Auth & route protection](./guides/auth-and-route-protection) |
 | Sandbox / workspace           | `agent/sandbox/`, [Sandbox](./sandbox)                                           |
-| Telemetry & debugging         | `agent/instrumentation.ts`, [Instrumentation](./observability/instrumentation)   |
+| Telemetry & debugging         | `agent/instrumentation/`, [Instrumentation](./observability/instrumentation)     |
 
 ## What to read next
 

package/docs/concepts/security-model.md

@@ -42,7 +42,10 @@ eve sends data where your agent configuration and runtime choices send it:
 - Model inputs and outputs flow to the model or routing path selected in `agent.ts`, such as a Vercel AI Gateway model id or a provider-authored `LanguageModel`.
 - Tool and connection calls flow to the external services, MCP servers, OpenAPI endpoints, and channels you configure.
 - Sandbox commands can reach network destinations allowed by the sandbox network policy.
-- Telemetry and eval data flows to the exporters and providers you configure in `instrumentation.ts` or eval settings.
+- Telemetry flows to destinations configured under `agent/instrumentation/`.
+  eve also records local traces during `eve dev` and exports to Vercel Agent
+  Runs in preview and production by default. Eval data flows to the reporters
+  configured in eval settings.
 
 eve stores durable session and workflow state needed to resume conversations, stream events, replay completed steps, and show run observability. You are responsible for deciding whether the selected channels, model providers, connected services, sandbox egress destinations, telemetry exporters, retention settings, and deletion controls are appropriate for your data and use case.
 

package/docs/evals/assertions.mdx

@@ -162,7 +162,7 @@ Every assertion returns a chainable handle. Severity rides on the assertion, so
 - `.atLeast(threshold)` is soft with a bar (equivalent to `.soft(threshold)`).
 - `.label(name)` adds a stable name when an eval records several assertions from the same family.
 
-The defaults are chosen so you rarely set severity. Run-level methods and `includes`/`equals`/`matches` are gates; `similarity` and every `t.judge.*` assertion are soft. Annotate only when you deviate:
+The defaults are chosen so you rarely set severity. Run-level methods and `includes`/`equals`/`matches` are gates; `similarity` and every `t.judge(...)` assertion are soft. Annotate only when you deviate:
 
 ```ts
 t.calledTool("get_weather").soft(); // record the tool call as a metric, don't gate

package/docs/evals/cases.mdx

@@ -70,7 +70,7 @@ export default defineEval({
   async test(t) {
     const draft = await t.send("Draft the follow-up email.");
     t.check(draft.message, includes("Best regards"));
-    t.judge.autoevals.closedQA("professional tone", { on: draft.message }).atLeast(0.6);
+    t.judge("professional tone", { on: draft.message }).atLeast(0.6);
 
     await draft.session.send("Now send it.");
     t.calledTool("send_email");
@@ -94,11 +94,9 @@ export default defineEval({
     t.succeeded();
     second.messageIncludes("marigold");
 
-    t.judge.autoevals
-      .closedQA("The assistant remembers the user's favorite word across turns", {
-        on: first.session.transcript,
-      })
-      .atLeast(0.8);
+    t.judge("The assistant remembers the user's favorite word across turns", {
+      on: first.session.transcript,
+    }).atLeast(0.8);
   },
 });
 ```

package/docs/evals/judge.mdx

@@ -1,9 +1,9 @@
 ---
 title: "Judge"
-description: "Grade evals with an LLM judge via t.judge.autoevals, set thresholds on the assertion, and configure the judge model."
+description: "Grade evals with evaluation models using criteria, typed questions, or batches, and set thresholds on each assertion."
 ---
 
-When no deterministic [assertion](./assertions) captures what "good" means (factual correctness, summary quality, free-form criteria), grade the run with an LLM judge. The `t.judge.*` assertions are the only model-backed ones, and they use a judge model that is resolved separately from the agent under test. eve only uses it for scoring, never to swap out the agent.
+Use `t.judge(...)` when a deterministic [assertion](./assertions) cannot capture what a good response looks like. It calls [`evaluate` from `eve/ai`](../guides/evaluate#evaluate-inside-a-tool) and defaults to `typesafe-ai/jev`. The judge is separate from the agent under test and is used only for scoring.
 
 ```ts
 import { defineEval } from "eve/evals";
@@ -12,110 +12,153 @@ export default defineEval({
   async test(t) {
     await t.send("Explain quantum tunneling to a 10-year-old.");
     t.succeeded();
-    t.judge.autoevals.closedQA("uses no math beyond arithmetic").atLeast(0.8);
+    t.judge("The response uses no math beyond arithmetic.").atLeast(0.8);
   },
 });
 ```
 
-## The graders
+No judge configuration is required. Configure credentials for the evaluator as described under [Configuring the judge model](#configuring-the-judge-model).
 
-The judges live under `t.judge.autoevals`. The namespace names the [Braintrust autoevals](https://github.com/braintrustdata/autoevals) grader family, so the factuality and closedQA semantics are autoevals', not eve-invented. Each grader scores the most recently settled turn’s assistant message by default and is soft by default (tracked, no gate):
+## The graders
 
-| Grader                                   | Grades                                                                                 |
-| ---------------------------------------- | -------------------------------------------------------------------------------------- |
-| `t.judge.autoevals.factuality(expected)` | Factual consistency of the reply against an expected answer (A–E buckets)              |
-| `t.judge.autoevals.summarizes(expected)` | How well the reply summarizes the expected text                                        |
-| `t.judge.autoevals.closedQA(criteria)`   | Whether the reply satisfies a free-form yes/no criterion (no expected answer to match) |
-| `t.judge.autoevals.sql(expected)`        | Semantic equivalence of two SQL statements                                             |
+`t.judge` accepts a criteria string, one typed question, or a batch of named questions. A string becomes a boolean question asking whether the response meets those criteria. A single question returns an assertion handle; a batch returns one handle per question.
 
-Judges receive the latest message text from the session whose turn most recently settled as the input prompt. Approval responses and stream reads preserve that prompt. For multipart messages, including `session.sendFile()`, the prompt contains the text parts.
+| Question                     | Assertion score                                                 |
+| ---------------------------- | --------------------------------------------------------------- |
+| Criteria string or `boolean` | The returned probability of true, from 0 to 1                   |
+| `score`                      | The returned rubric position divided by the highest level index |
+| `choice`                     | 1 when the selected option equals `expected`; otherwise 0       |
 
-A failed `closedQA` assertion prints the input and judge explanation:
+Boolean scores are model estimates, not guaranteed calibrated confidence. A rubric needs at least two levels, ordered from worst to best. The SDK can return a fractional position on that rubric; eve normalizes it to the 0–1 scale used by assertion thresholds.
 
-```text
-✗ judge.autoevals.closedQA [citation] (0% < 80%): prompt: "Name the source."
-  criteria: "cites a source"
-  response: "It is widely believed."
-  rationale: "The response does not cite a source."
-  choice: "N"
+```ts
+t.judge({
+  type: "score",
+  instructions: "Grade the response's clarity.",
+  criteria: ["Unclear", "Mostly clear", "Clear and concise"],
+})
+  .label("clarity")
+  .atLeast(0.75);
 ```
 
-The reference or criteria is the positional argument. An options object follows:
+Choice questions require an `expected` option key. eve uses that expectation locally for scoring; it does not send it to the evaluator. Instructions and rubric descriptions accept strings, JSON objects, or JSON arrays. Descriptions can also be `null`.
+
+## Choose the value to grade
 
-- `on` is the value to grade, defaulting to the most recently settled turn’s assistant message. Pass an intermediate draft or parsed value to grade it instead.
-- `model` and `modelOptions` are a per-call judge override (see below).
+By default, the evaluator receives `{ input, output }`: the latest prompt and assistant message from the session whose turn most recently settled. Approval responses and stream reads preserve that prompt. For multipart messages, including `session.sendFile()`, the prompt contains the text parts.
+
+Pass `on` to replace the output. JSON values stay structured rather than being converted to strings:
 
 ```ts
 const draft = await t.send("Draft the welcome email.");
-t.judge.autoevals.closedQA("professional tone", { on: draft.message }).atLeast(0.6);
+t.judge("The response has a professional tone.", { on: draft.message }).atLeast(0.6);
 ```
 
-For a multi-turn eval, pass `session.transcript` to grade a session's complete observed conversation instead of only its final reply:
+For a multi-turn eval, pass `session.transcript`:
 
 ```ts
 const { session } = await t.send("My favorite word is marigold. Remember it.");
 await session.send("What is my favorite word?");
+t.judge("The assistant remembers the user's favorite word across turns.", {
+  on: session.transcript,
+}).atLeast(0.8);
+```
+
+The transcript contains user and assistant messages in turn order, excluding reasoning, tool calls, and tool results. See [Multi-turn evals](./cases#multi-turn-evals).
 
-t.judge.autoevals
-  .closedQA("The assistant remembers the user's favorite word across turns", {
-    on: session.transcript,
-  })
-  .atLeast(0.8);
+## Evaluate several questions together
+
+A batch makes one evaluation request against one shared state. Pass `state` to replace the default `{ input, output }` completely. Batch calls use `state` instead of `on`.
+
+```ts
+const judgments = t.judge({
+  state: { response: turn.message ?? "", reference },
+  questions: {
+    accurate: {
+      type: "boolean",
+      instructions: "Is the response consistent with the reference?",
+    },
+    outcome: {
+      type: "choice",
+      instructions: "Classify the response.",
+      criteria: {
+        answered: "Answers the request",
+        declined: "Declines the request",
+      },
+      expected: "answered",
+    },
+  },
+});
+judgments.accurate.atLeast(0.9);
+judgments.outcome.gate();
 ```
 
-`session.transcript` contains the session's user and assistant messages in turn order. It excludes reasoning, tool calls, and tool results. See [Multi-turn evals](./cases#multi-turn-evals) for the transcript format and independent sessions.
+Keys identify questions and their assertion handles; instructions must explain what to judge. Every question sees the same state. Use separate calls for unrelated inputs. A batch must contain at least one question, and a provider or validation failure fails every assertion in the batch.
 
 ## Soft scoring and thresholds
 
-Judge assertions are soft, so the threshold rides on the assertion handle. There is no separate thresholds map:
+Judges are soft by default:
 
-- **No threshold** is tracked-only. The score lands in reports and artifacts and never fails the eval. Use it to watch a metric without gating on it.
-- `.atLeast(threshold)` is a soft bar. A below-threshold score marks the eval `scored`, fatal only under `eve eval --strict`.
-- `.gate(threshold)` promotes a judge to a hard gate that fails the eval outright.
+- Without a threshold, the score is tracked in reports and artifacts without failing the eval.
+- `.atLeast(threshold)` sets a soft bar. A lower score marks the eval `scored`, fatal only under `eve eval --strict`.
+- `.gate(threshold)` makes a lower score fail the eval outright. An omitted gate threshold is 1.
 
 ```ts
-t.judge.autoevals.closedQA("cites a source"); // tracked, never fails
-t.judge.autoevals.closedQA("cites a source").label("citation").atLeast(0.6);
-t.judge.autoevals.factuality(reference).gate(0.8); // hard gate at 0.8
+t.judge("The response cites a source.");
+t.judge("The response cites a source.").label("citation").atLeast(0.8);
+t.judge("The response cites a source.").gate(0.9);
 ```
 
-A judge runs once per assertion and burns tokens, so reach for one only when nothing deterministic will do. Judge calls start when recorded, and the runner waits for all of them during finalization; assertion handles themselves are intentionally not awaitable.
+A boolean probability need not reach exactly 1 even for a good response, so choose an explicit gate threshold for boolean judgments. Choice scores are binary.
+
+Calls capture their state and questions and start immediately. The runner waits for them during finalization; handles are not awaitable. Each standalone call makes its own request, while a batch shares one request. Eval timeouts cancel pending judgments and stop waiting even if a provider ignores cancellation.
 
 ## Configuring the judge model
 
-The judge model is resolved once when the runner builds `t`. It is **never** the model under test. Three levels resolve innermost-wins:
+Model settings resolve in this order:
 
-1. **Per-call**: `t.judge.autoevals.closedQA("…", { model, modelOptions })`.
-2. **Per-eval**: `defineEval({ judge: { model, modelOptions }, test })`.
-3. **Project default**: `defineEvalConfig({ judge: { model, modelOptions } })` in `evals.config.ts`.
+1. Per-call: `t.judge("…", { model, modelOptions })`.
+2. Per-eval: `defineEval({ judge: { model, modelOptions }, test })`.
+3. Project default: `defineEvalConfig({ judge: { model, modelOptions } })`.
+4. The default model used by `evaluate`, currently `typesafe-ai/jev`.
+
+A per-eval `judge` replaces the project's judge configuration. Per-call fields override the resolved configuration. `judge` and its `model` are optional; deterministic evals make no evaluation requests.
 
 ```ts title="evals/evals.config.ts"
 import { defineEvalConfig } from "eve/evals";
 
 export default defineEvalConfig({
-  judge: { model: "openai/gpt-5.6-luna" }, // the default judge for every eval in this tree
+  judge: { model: "typesafe-ai/jev" },
 });
 ```
 
-```ts title="evals/quantum.eval.ts"
-import { defineEval } from "eve/evals";
+String IDs resolve through the AI SDK default provider, or Vercel AI Gateway when none is configured. Gateway uses `AI_GATEWAY_API_KEY` or Vercel OIDC credentials; eve's existing local Gateway connection is also honored by `evaluate`. An explicit provider evaluation model uses that provider's credentials:
 
-export default defineEval({
-  judge: { model: "anthropic/claude-opus-4.8" }, // a stronger judge for this eval
-  async test(t) {
-    await t.send("Explain quantum tunneling to a 10-year-old.");
-    t.judge.autoevals.factuality(reference).atLeast(0.7);
-    t.judge.autoevals.closedQA("is concise", { model: "anthropic/claude-haiku-4.5" }); // cheaper, per-call
-  },
+```ts
+import { openai } from "@ai-sdk/openai";
+
+t.judge("The response addresses the request.", {
+  model: openai.evaluationModel("gpt-5.6-luna"),
+  modelOptions: { providerOptions: { openai: { reasoningEffort: "high" } } },
 });
 ```
 
-`judge` in `evals.config.ts` is optional, and a tree of fully deterministic evals can omit it. Calling `t.judge.*` with no judge model resolved records a failed gate: the runner scores the assertion after the `test` function runs, the missing model throws, and the eval fails with that message.
+Use the provider's `evaluationModel` factory, not a language model instance. AI SDK supplies evaluation adapters for supported language models. Provider support and the evaluation contract are experimental; see [AI SDK evaluation](https://ai-sdk.dev/docs/ai-sdk-core/evaluation).
+
+String model IDs resolve through the default provider's evaluation API. Gateway accepts native evaluation models such as `typesafe-ai/jev`; passing a language model ID such as `openai/gpt-5.6-luna` does not automatically select the OpenAI adapter. Use an evaluation-model instance as shown above for language models.
+
+Missing credentials, invalid answers, unsupported question types, and provider errors become failed gates even for tracked-only judgments. They do not silently skip the eval or select another model. Validation, authentication, and retries use the shared `evaluate` implementation.
+
+## Diagnostics and migration
+
+Reports use `judge.boolean`, `judge.score`, and `judge.choice`; batch assertion names also include their question key. Diagnostics include the captured state, question, and raw answer. Assertion metadata retains the normalized score, model identity, token usage, warnings, and available provider and response metadata. Usage for a batch is marked as shared across its assertions, not a separate charge for each question. Evaluation models do not promise a prose rationale.
+
+The autoevals graders have been removed. Replace `t.judge.autoevals.closedQA(criteria, options)` with `t.judge(criteria, options)`. Scores now represent a probability rather than a binary yes/no answer, so review thresholds against your examples. For factuality, summaries, or SQL equivalence, write an explicit question and include the reference in shared state. No old grader buckets are preserved.
 
-A **string model id** (e.g. `"anthropic/claude-opus-4.8"`) routes through the Vercel AI Gateway and needs `AI_GATEWAY_API_KEY` or `VERCEL_OIDC_TOKEN` in the environment. An **AI SDK `LanguageModel` instance** is used directly. With a model configured but no credentials, a judge-backed eval **skips visibly** rather than failing, so the run reports the skip instead of a spurious error. For provider-specific judge settings, use `modelOptions.providerOptions`.
+Change configured language model instances to evaluation model instances. `modelOptions.providerOptions`, assertion labels, and threshold methods remain available. [Braintrust reporting](./reporters) continues to accept judge scores independently of the grader implementation.
 
 ## What to read next
 
 - [Assertions](./assertions): deterministic run-level and value assertions
-- [Reporters](./reporters): ship judged scores to Braintrust experiments
-- [Targets](./targets): local vs remote targets for judge-backed evals
+- [Reporters](./reporters): publish scores and inspect results
+- [Targets](./targets): local and remote eval targets

package/docs/evals/overview.mdx

@@ -51,12 +51,12 @@ import { defineEvalConfig } from "eve/evals";
 import { Braintrust } from "eve/evals/reporters";
 
 export default defineEvalConfig({
-  judge: { model: "openai/gpt-5.6-luna" },
+  judge: { model: "typesafe-ai/jev" },
   reporters: [Braintrust({ projectName: "my-agent" })],
 });
 ```
 
-Everything is optional. `judge` sets the default model for [LLM-as-judge](./judge) assertions (`t.judge.*`); a tree of fully deterministic evals can omit it. `reporters`, `maxConcurrency`, and `timeoutMs` round out the defaults. Config `reporters` observe every eval in the run, so set one `Braintrust()` here instead of adding it to each eval. CLI flags (`--max-concurrency`, `--timeout`) and per-eval values take precedence over the config defaults.
+Everything is optional. `judge` sets the default model for [LLM-as-judge](./judge) assertions (`t.judge(...)`); omit it to use the default evaluator, `typesafe-ai/jev`. Deterministic evals make no judge requests. `reporters`, `maxConcurrency`, and `timeoutMs` round out the defaults. Config `reporters` observe every eval in the run, so set one `Braintrust()` here instead of adding it to each eval. CLI flags (`--max-concurrency`, `--timeout`) and per-eval values take precedence over the config defaults.
 
 ## Deterministic fixture models
 
@@ -109,22 +109,24 @@ Each surface matches a genuinely different kind of judgment:
 
 - **Scoped methods** read the final whole run on `t`, snapshot one independent session when invoked there, or inspect one immutable `EveEvalTurn`. See [Assertions](./assertions).
 - **`t.check(value, assertion)`** grades an explicit value with a deterministic builder from `eve/evals/expect`, such as `t.check(turn.message, includes("sunny"))`. Grade `turn.message`, an intermediate draft, parsed JSON, or anything else. See [Assertions](./assertions).
-- **`t.judge.autoevals.*`** is the LLM-as-judge surface, like `t.judge.autoevals.closedQA("cites a source")`. It grades the most recently settled turn’s assistant message by default; pass `{ on: turn.session.transcript }` to grade a multi-turn conversation. The judge uses the configured judge model, never the agent under test. See [Judge](./judge).
+- **`t.judge(...)`** is the LLM-as-judge surface, like `t.judge("cites a source")`. It grades the most recently settled turn’s assistant message by default; pass `{ on: turn.session.transcript }` to grade a multi-turn conversation. The judge uses the configured judge model, never the agent under test. See [Judge](./judge).
 
 ## Gate vs soft
 
 Every assertion returns a chainable handle, so severity rides on the assertion itself. There is no separate thresholds map.
 
 - **Gates** are hard. A failed gate marks the eval `failed` and `eve eval` exits non-zero. Run-level methods, `includes`, `equals`, and `matches` are gates by default.
-- **Soft** assertions are tracked data. They land in reports and artifacts, and a below-threshold soft assertion marks the eval `scored` (visible but not fatal, unless you pass `--strict`). `similarity` and every `t.judge.*` assertion are soft by default. A soft assertion with no threshold is tracked-only and never fails.
+- **Soft** assertions are tracked data. They land in reports and artifacts, and a below-threshold soft assertion marks the eval `scored` (visible but not fatal, unless you pass `--strict`). `similarity` and every `t.judge(...)` assertion are soft by default. A soft assertion with no threshold is tracked-only and never fails.
 
 Override per assertion: `.gate(threshold?)` promotes to a hard gate, `.soft(threshold?)` demotes to tracked, and `.atLeast(threshold)` is a soft assertion with a bar.
 
 ```ts
 t.succeeded(); // gate
 t.calledTool("get_weather").soft(); // record as a metric, don't gate
-t.judge.autoevals.closedQA("cites a source"); // soft, tracked (no threshold)
-t.judge.autoevals.factuality(reference).atLeast(0.7); // soft, gated under --strict at 0.7
+t.judge("cites a source"); // soft, tracked (no threshold)
+t.judge("The response is consistent with output.reference.", {
+  on: { response: turn.message ?? "", reference },
+}).atLeast(0.7); // soft, gated under --strict at 0.7
 ```
 
 Use `await t.require(value, assertion)` for a gate that must pass before the script can safely continue. Use `t.skip(reason)` as the first operation for an intentionally unsupported target capability.

package/docs/evals/reporters.mdx

@@ -18,7 +18,7 @@ import { defineEvalConfig } from "eve/evals";
 import { Braintrust } from "eve/evals/reporters";
 
 export default defineEvalConfig({
-  judge: { model: "openai/gpt-5.6-luna" },
+  judge: { model: "typesafe-ai/jev" },
   reporters: [Braintrust({ projectName: "weather-agent" })],
 });
 ```

package/docs/guides/dev-tui.md

@@ -102,7 +102,7 @@ By default, the UI shows `stderr` logs. Use `/loglevel <all|stderr|sandbox|none>
 
 Every `eve dev` process writes diagnostic logs to `.eve/logs/`, regardless of the display mode. Read them with [`eve logs`](../reference/cli#eve-logs).
 
-Use `/traces` to inspect traces recorded during local development. See [Instrumentation](../observability/instrumentation#local-traces) for trace capture and retention settings.
+Use `/traces` to inspect traces recorded during local development. See [Local traces](../observability/otel#local-traces) for trace capture and retention settings.
 
 ## Display options