npm - @cursor/july - Versions diffs - 0.1.5 → 0.1.7 - Mend

@cursor/july 0.1.5 → 0.1.7

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (177) hide show

package/dist/ab.d.ts +8 -95
package/dist/ab.d.ts.map +1 -1
package/dist/ab.js +9 -150
package/dist/bin/agent-serve.js +41 -8
package/dist/channels/slack/post-update-delivery.d.ts +85 -0
package/dist/channels/slack/post-update-delivery.d.ts.map +1 -0
package/dist/docs/404.html +2 -2
package/dist/docs/ab.html +4 -4
package/dist/docs/assets/{app.DabPG-io.js → app.COTN7wgo.js} +1 -1
package/dist/docs/assets/chunks/@localSearchIndexroot.B7UcKvIn.js +1 -0
package/dist/docs/assets/chunks/{VPLocalSearchBox.jmyr0bU0.js → VPLocalSearchBox.BW3TBdT0.js} +1 -1
package/dist/docs/assets/chunks/{theme.DysN9-VN.js → theme.BEJW0vE7.js} +2 -2
package/dist/docs/assets/deployment.md.BtfEsc9S.js +55 -0
package/dist/docs/assets/deployment.md.BtfEsc9S.lean.js +1 -0
package/dist/docs/assets/example-agents_approval-buddy.md.8R5phXb5.js +10 -0
package/dist/docs/assets/example-agents_approval-buddy.md.8R5phXb5.lean.js +1 -0
package/dist/docs/assets/example-agents_benny.md.B0gjhI-p.js +7 -0
package/dist/docs/assets/example-agents_benny.md.B0gjhI-p.lean.js +1 -0
package/dist/docs/assets/example-agents_bugbot.md.DelIdhxB.js +11 -0
package/dist/docs/assets/example-agents_bugbot.md.DelIdhxB.lean.js +1 -0
package/dist/docs/assets/example-agents_codebase-wiki.md.DC6sgwn0.js +8 -0
package/dist/docs/assets/example-agents_codebase-wiki.md.DC6sgwn0.lean.js +1 -0
package/dist/docs/assets/example-agents_codeowners-review.md.Ku_tG2RY.js +8 -0
package/dist/docs/assets/example-agents_codeowners-review.md.Ku_tG2RY.lean.js +1 -0
package/dist/docs/assets/example-agents_concierge.md.4rQTSMXt.js +23 -0
package/dist/docs/assets/example-agents_concierge.md.4rQTSMXt.lean.js +1 -0
package/dist/docs/assets/example-agents_fsd.md.CzgUrDfi.js +15 -0
package/dist/docs/assets/example-agents_fsd.md.CzgUrDfi.lean.js +1 -0
package/dist/docs/assets/example-agents_index.md.CRqJlnIf.js +2 -0
package/dist/docs/assets/example-agents_index.md.CRqJlnIf.lean.js +1 -0
package/dist/docs/assets/example-agents_knowledge-base.md.BPJiVueF.js +11 -0
package/dist/docs/assets/example-agents_knowledge-base.md.BPJiVueF.lean.js +1 -0
package/dist/docs/assets/example-agents_security-reviewer.md.D2rtwDTO.js +19 -0
package/dist/docs/assets/example-agents_security-reviewer.md.D2rtwDTO.lean.js +1 -0
package/dist/docs/assets/example-agents_slack-agent.md.buLbgvBf.js +5 -0
package/dist/docs/assets/example-agents_slack-agent.md.buLbgvBf.lean.js +1 -0
package/dist/docs/assets/example-agents_weather-agent.md.C9Qv-W0o.js +24 -0
package/dist/docs/assets/example-agents_weather-agent.md.C9Qv-W0o.lean.js +1 -0
package/dist/docs/assets/index.md.COiu-1jL.js +20 -0
package/dist/docs/assets/{index.md.Cylk70gg.lean.js → index.md.COiu-1jL.lean.js} +1 -1
package/dist/docs/assets/reference_cli.md.D189RBCH.js +60 -0
package/dist/docs/assets/reference_cli.md.D189RBCH.lean.js +1 -0
package/dist/docs/building-with-agents.html +4 -4
package/dist/docs/concepts.html +4 -4
package/dist/docs/deployment.html +58 -17
package/dist/docs/evals.html +4 -4
package/dist/docs/example-agents/approval-buddy.html +34 -0
package/dist/docs/example-agents/benny.html +31 -0
package/dist/docs/example-agents/bugbot.html +35 -0
package/dist/docs/example-agents/codebase-wiki.html +32 -0
package/dist/docs/example-agents/codeowners-review.html +32 -0
package/dist/docs/example-agents/concierge.html +47 -0
package/dist/docs/example-agents/fsd.html +39 -0
package/dist/docs/example-agents/index.html +26 -0
package/dist/docs/example-agents/knowledge-base.html +35 -0
package/dist/docs/example-agents/security-reviewer.html +43 -0
package/dist/docs/example-agents/slack-agent.html +29 -0
package/dist/docs/example-agents/weather-agent.html +48 -0
package/dist/docs/guides/agent-to-agent.html +4 -4
package/dist/docs/guides/cloud-runtime.html +5 -5
package/dist/docs/guides/github.html +4 -4
package/dist/docs/guides/human-in-the-loop.html +4 -4
package/dist/docs/guides/slack.html +4 -4
package/dist/docs/guides/webhooks.html +4 -4
package/dist/docs/hashmap.json +1 -1
package/dist/docs/hillclimbing.html +4 -4
package/dist/docs/index.html +7 -7
package/dist/docs/quickstart.html +4 -4
package/dist/docs/reference/agent-config.html +4 -4
package/dist/docs/reference/channels.html +4 -4
package/dist/docs/reference/cli.html +52 -30
package/dist/docs/reference/connections.html +4 -4
package/dist/docs/reference/hooks.html +4 -4
package/dist/docs/reference/http-api.html +4 -4
package/dist/docs/reference/instructions.html +4 -4
package/dist/docs/reference/playground.html +4 -4
package/dist/docs/reference/project-layout.html +4 -4
package/dist/docs/reference/schedules.html +4 -4
package/dist/docs/reference/sessions.html +4 -4
package/dist/docs/reference/skills.html +4 -4
package/dist/docs/reference/subagents.html +4 -4
package/dist/docs/reference/tools.html +4 -4
package/dist/docs/scaffolding-agents.html +4 -4
package/dist/docs/storage.html +4 -4
package/dist/docs/troubleshooting.html +4 -4
package/dist/evals.d.ts +5 -62
package/dist/evals.d.ts.map +1 -1
package/dist/evals.js +3 -66
package/dist/index.d.ts +1 -1
package/dist/index.d.ts.map +1 -1
package/dist/internal/ab-collector.d.ts +7 -5
package/dist/internal/ab-collector.d.ts.map +1 -1
package/dist/internal/ab-collector.js +3 -14
package/dist/internal/ab-snapshot.d.ts +2 -4
package/dist/internal/ab-snapshot.d.ts.map +1 -1
package/dist/internal/cli-ax.d.ts +33 -5
package/dist/internal/cli-ax.d.ts.map +1 -1
package/dist/internal/cli-ax.js +428 -87
package/dist/internal/cli-deploy.js +1 -1
package/dist/internal/discovery.js +3 -3
package/dist/internal/eval-run-store.d.ts +35 -30
package/dist/internal/eval-run-store.d.ts.map +1 -1
package/dist/internal/eval-run-store.js +88 -100
package/dist/internal/evals-client.d.ts +96 -0
package/dist/internal/evals-client.d.ts.map +1 -0
package/dist/internal/evals-client.js +262 -0
package/dist/internal/init-project.d.ts.map +1 -1
package/dist/internal/init-project.js +1 -0
package/dist/internal/persistence-coordinator.d.ts +127 -0
package/dist/internal/persistence-coordinator.d.ts.map +1 -0
package/dist/internal/playground-proxy.d.ts +5 -5
package/dist/internal/playground-proxy.js +3 -3
package/dist/internal/resolve-prod-target.d.ts +30 -0
package/dist/internal/resolve-prod-target.d.ts.map +1 -1
package/dist/internal/resolve-prod-target.js +74 -2
package/dist/internal/server.d.ts.map +1 -1
package/dist/internal/server.js +16 -5
package/dist/internal/session-engine.d.ts +1 -2
package/dist/internal/session-engine.d.ts.map +1 -1
package/dist/internal/session-engine.js +14 -31
package/dist/internal/storage-coordinator.d.ts +16 -15
package/dist/internal/storage-coordinator.d.ts.map +1 -1
package/dist/internal/storage-coordinator.js +73 -80
package/dist/persistence.d.ts +184 -0
package/dist/persistence.d.ts.map +1 -0
package/dist/playground/assets/cursor-icons-16-CQ50JpfO.woff2 +0 -0
package/dist/playground/assets/index-72vCOBWO.js +86 -0
package/dist/playground/assets/index-BjnMwYoR.css +1 -0
package/dist/playground/index.html +2 -2
package/dist/storage.d.ts +51 -10
package/dist/storage.d.ts.map +1 -1
package/dist/storage.js +27 -10
package/docs/README.md +34 -5
package/docs/deployment.md +352 -149
package/docs/example-agents/approval-buddy.md +270 -0
package/docs/example-agents/benny.md +186 -0
package/docs/example-agents/bugbot.md +231 -0
package/docs/example-agents/codebase-wiki.md +174 -0
package/docs/example-agents/codeowners-review.md +195 -0
package/docs/example-agents/concierge.md +205 -0
package/docs/example-agents/fsd.md +330 -0
package/docs/example-agents/index.md +102 -0
package/docs/example-agents/knowledge-base.md +171 -0
package/docs/example-agents/security-reviewer.md +296 -0
package/docs/example-agents/slack-agent.md +146 -0
package/docs/example-agents/weather-agent.md +302 -0
package/docs/reference/cli.md +546 -147
package/package.json +1 -1
package/src/ab.ts +9 -261
package/src/bin/agent-serve.ts +46 -7
package/src/evals.ts +5 -119
package/src/index.ts +2 -0
package/src/internal/ab-collector.ts +12 -22
package/src/internal/ab-snapshot.ts +2 -4
package/src/internal/cli-ax.ts +551 -104
package/src/internal/cli-deploy.ts +1 -1
package/src/internal/discovery.ts +2 -2
package/src/internal/eval-run-store.ts +91 -100
package/src/internal/evals-client.ts +431 -0
package/src/internal/init-project.ts +1 -0
package/src/internal/playground-proxy.ts +5 -5
package/src/internal/resolve-prod-target.ts +101 -3
package/src/internal/server.ts +17 -3
package/src/internal/session-engine.ts +9 -29
package/src/internal/storage-coordinator.ts +109 -101
package/src/storage.ts +79 -14
package/dist/docs/assets/chunks/@localSearchIndexroot.QwK5BtEH.js +0 -1
package/dist/docs/assets/deployment.md.DTKwE15Z.js +0 -14
package/dist/docs/assets/deployment.md.DTKwE15Z.lean.js +0 -1
package/dist/docs/assets/index.md.Cylk70gg.js +0 -20
package/dist/docs/assets/reference_cli.md.Bv6pOxcF.js +0 -38
package/dist/docs/assets/reference_cli.md.Bv6pOxcF.lean.js +0 -1
package/dist/internal/json-dir-store.js +0 -100
package/dist/playground/assets/cursor-icons-outline-BxTT_FVJ.woff2 +0 -0
package/dist/playground/assets/index-BEauYlII.css +0 -1
package/dist/playground/assets/index-BtM0wEGg.js +0 -319
package/src/internal/json-dir-store.ts +0 -109

package/dist/playground/index.html CHANGED Viewed

@@ -8,8 +8,8 @@
     />
     <meta name="viewport" content="width=device-width, initial-scale=1" />
     <title>agent-serve playground</title>
-    <script type="module" crossorigin src="./assets/index-BtM0wEGg.js"></script>
-    <link rel="stylesheet" crossorigin href="./assets/index-BEauYlII.css">
+    <script type="module" crossorigin src="./assets/index-72vCOBWO.js"></script>
+    <link rel="stylesheet" crossorigin href="./assets/index-BjnMwYoR.css">
   </head>
   <body>
     <div id="root"></div>

package/dist/storage.d.ts CHANGED Viewed

@@ -4,10 +4,11 @@
  * Author `agent/storage.ts` with {@link defineStorage} to mirror the
  * framework's durable state into storage you own (a database, S3, a data
  * pipeline, …). Without it, state lives under `--state-root` on local disk
- * only (and eval/A/B history follows the narrower `persistRuns` /
- * `persistSamples` / `persistSnapshots` hooks).
+ * only, and eval/A/B history stays in process memory.
  *
- * The sink is a plain key-value store — four functions, no schema:
+ * The sink is a plain key-value store — four functions, no schema — plus
+ * two optional dedicated tables ({@link StorageConfig.evals} and
+ * {@link StorageConfig.abs}) for eval-run and A/B history:
  *
  * ```ts
  * import { defineStorage } from "@anysphere/agent-serve/storage";
@@ -17,6 +18,8 @@
  *   get: (key) => db.get(key),
  *   delete: (key) => db.delete(key),
  *   list: (prefix) => db.listByPrefix(prefix), // [{ key, value }] in key order
+ *   evals: { put: ..., delete: ..., list: ... }, // eval-runs table
+ *   abs: { putSample: ... },                     // A/B table
  * });
  * ```
  *
@@ -39,6 +42,9 @@
  * event log under `--state-root` remains the live source of truth; this
  * interface is the durable mirror.
  */
+import type { ABMetricSample } from "./ab.js";
+import type { EvalRunSnapshot } from "./evals.js";
+import type { ABSnapshot } from "./internal/ab-snapshot.js";
 import type { JsonValue } from "./types.js";
 /** One `{ key, value }` pair returned by {@link StorageConfig.list}. */
 export interface StorageEntry {
@@ -98,11 +104,50 @@ export interface StorageRestorePolicy {
      */
     maxTotalBytes?: number;
 }
+/**
+ * Dedicated table for playground eval batches, keyed by `runId`. The
+ * framework saves a full {@link EvalRunSnapshot} as a batch starts,
+ * progresses, and finishes; prunes runs past the playground history
+ * window; and lists everything back at serve start.
+ */
+export interface StorageEvalsTable {
+    /** Upsert one run snapshot under `run.runId` (last write wins). */
+    put(run: EvalRunSnapshot, ctx: StorageContext): void | Promise<void>;
+    /** Remove a run pruned out of the playground history window. */
+    delete(runId: string, ctx: StorageContext): void | Promise<void>;
+    /** All saved runs (any order). Called once when the serve process hydrates. */
+    list(ctx: StorageContext): EvalRunSnapshot[] | Promise<EvalRunSnapshot[]>;
+}
+/**
+ * Dedicated table for live A/B metrics: one row per cumulative metric
+ * sample (unique per `experiment` + `sessionId` + `at` — one boundary
+ * event emits a sample per enrolled experiment), plus — optionally — the latest
+ * aggregate snapshot so a replacement host with no local sessions can
+ * still serve the A/Bs surface.
+ */
+export interface StorageABTable {
+    /** Append one cumulative metric sample (turn complete/fail). */
+    putSample(sample: ABMetricSample, ctx: StorageContext): void | Promise<void>;
+    /** Upsert the latest aggregate snapshot (framework-throttled). Optional. */
+    putSnapshot?(snapshot: ABSnapshot, ctx: StorageContext): void | Promise<void>;
+    /** Latest saved aggregate snapshot (replacement-host backfill). Optional. */
+    getSnapshot?(ctx: StorageContext): ABSnapshot | undefined | null | Promise<ABSnapshot | undefined | null>;
+}
 export interface StorageConfig {
     /** Optional label surfaced on `GET /v1/info` diagnostics. */
     name?: string;
     /** Timing knobs; see {@link StoragePolicy}. */
     policy?: StoragePolicy;
+    /**
+     * Dedicated eval-runs table. Omit it and playground eval history stays
+     * in process memory (lost on restart).
+     */
+    evals?: StorageEvalsTable;
+    /**
+     * Dedicated A/B table. Omit it and metric samples are not exported
+     * (session event logs remain the assignment/fold source of truth).
+     */
+    abs?: StorageABTable;
     /**
      * Store one value under a key (upsert, last-write-wins). Called on the
      * framework's schedule — never concurrently, always in order. Keep it
@@ -160,9 +205,9 @@ export declare const STORAGE_KEY_ROOT: "agentkit/v1";
  * | `agentkit/v1/{agent}/session-events/{sessionId}/{index}` | `SessionEvent[]` chunk (index = first event's index, zero-padded) |
  * | `agentkit/v1/{agent}/continuation/{channelId}/{token}` | `{ sessionId }` |
  * | `agentkit/v1/{agent}/reminder/{reminderId}` | `ReminderRecord` |
- * | `agentkit/v1/{agent}/eval-run/{runId}` | `EvalRunSnapshot` |
- * | `agentkit/v1/{agent}/ab-sample/{sessionId}/{at}` | `ABMetricSample` |
- * | `agentkit/v1/{agent}/ab-snapshot` | latest aggregate `ABSnapshot` |
+ *
+ * Eval-run and A/B history do not flow through this KV scheme — they have
+ * dedicated tables ({@link StorageConfig.evals} / {@link StorageConfig.abs}).
  */
 export declare const storageKeys: {
     readonly session: (agent: string, sessionId: string) => string;
@@ -172,10 +217,6 @@ export declare const storageKeys: {
     readonly continuation: (agent: string, channelId: string, continuationKey: string) => string;
     readonly reminder: (agent: string, reminderId: string) => string;
     readonly reminderPrefix: (agent: string) => string;
-    readonly evalRun: (agent: string, runId: string) => string;
-    readonly evalRunPrefix: (agent: string) => string;
-    readonly abSample: (agent: string, sessionId: string, at: string) => string;
-    readonly abSnapshot: (agent: string) => string;
 };
 /**
  * Max bytes a caller-controlled key segment may occupy after URI-encoding.

package/dist/storage.d.ts.map CHANGED Viewed

	@@ -1 +1 @@
1	- {"version":3,"file":"storage.d.ts","sourceRoot":"","sources":["../src/storage.ts"],"names":[],"mappings":"AAAA~~;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAwCG~~;~~AAIH~~,OAAO,KAAK,EAAE,SAAS,EAAE,MAAM,YAAY,CAAC;AAM5C,wEAAwE;AACxE,MAAM,WAAW,YAAY;IAC3B,GAAG,EAAE,MAAM,CAAC;IACZ,KAAK,EAAE,SAAS,CAAC;CAClB;AAED,yCAAyC;AACzC,MAAM,WAAW,cAAc;IAC7B,uEAAuE;IACvE,SAAS,EAAE,MAAM,CAAC;IAClB,sEAAsE;IACtE,WAAW,EAAE,MAAM,CAAC;IACpB;;;;;OAKG;IACH,MAAM,EAAE,QAAQ,GAAG,UAAU,GAAG,SAAS,CAAC;CAC3C;AAED,MAAM,WAAW,aAAa;IAC5B;;;;;;OAMG;IACH,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB;;;;;;;;;OASG;IACH,OAAO,CAAC,EAAE,oBAAoB,GAAG,KAAK,CAAC;CACxC;AAED,mFAAmF;AACnF,MAAM,WAAW,oBAAoB;IACnC,wFAAwF;IACxF,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB;;;;OAIG;IACH,QAAQ,CAAC,EAAE,MAAM,CAAC;IAClB;;;;;OAKG;IACH,aAAa,CAAC,EAAE,MAAM,CAAC;CACxB;AAED,MAAM,WAAW,aAAa;IAC5B,6DAA6D;IAC7D,IAAI,CAAC,EAAE,MAAM,CAAC;IACd,+CAA+C;IAC/C,MAAM,CAAC,EAAE,aAAa,CAAC;IACvB;;;;;;OAMG;IACH,GAAG,CAAC,GAAG,EAAE,MAAM,EAAE,KAAK,EAAE,SAAS,EAAE,GAAG,EAAE,cAAc,GAAG,IAAI,GAAG,OAAO,CAAC,IAAI,CAAC,CAAC;IAC9E,kEAAkE;IAClE,MAAM,CAAC,CAAC,GAAG,EAAE,MAAM,EAAE,GAAG,EAAE,cAAc,GAAG,IAAI,GAAG,OAAO,CAAC,IAAI,CAAC,CAAC;IAChE;;;;;;;OAOG;IACH,GAAG,CAAC,CACF,GAAG,EAAE,MAAM,EACX,GAAG,EAAE,cAAc,GAClB,SAAS,GAAG,SAAS,GAAG,IAAI,GAAG,OAAO,CAAC,SAAS,GAAG,SAAS,GAAG,IAAI,CAAC,CAAC;IACxE;;;;OAIG;IACH,IAAI,CAAC,CACH,MAAM,EAAE,MAAM,EACd,GAAG,EAAE,cAAc,GAClB,YAAY,EAAE,GAAG,OAAO,CAAC,YAAY,EAAE,CAAC,CAAC;CAC7C;AAED,MAAM,MAAM,iBAAiB,GAAG,aAAa,GAAG;IAC9C,QAAQ,CAAC,YAAY,EAAE,SAAS,CAAC;CAClC,CAAC;AAEF;;;GAGG;AACH,wBAAgB,aAAa,CAAC,MAAM,EAAE,aAAa,GAAG,iBAAiB,~~CAetE~~;AAMD;;;GAGG;AACH,eAAO,MAAM,gBAAgB,EAAG,aAAsB,CAAC;AAEvD;;;;;;;;;;;;;;;;;;;;;GAqBG;AACH,eAAO,MAAM,WAAW;aACtB,OAAO,UAAU,MAAM,aAAa,MAAM,KAAG,MAAM;aAEnD,aAAa,UAAU,MAAM,KAAG,MAAM;aAEtC,aAAa,UACJ,MAAM,aACF,MAAM,cACL,MAAM,KACjB,MAAM;aAET,mBAAmB,UAAU,MAAM,aAAa,MAAM,KAAG,MAAM;aAE/D,YAAY,UACH,MAAM,aACF,MAAM,mBACA,MAAM,KACtB,MAAM;aAET,QAAQ,UAAU,MAAM,cAAc,MAAM,KAAG,MAAM;aAErD,cAAc,UAAU,MAAM,KAAG,MAAM;~~aAEvC,OAAO,UAAU,MAAM,SAAS,MAAM,KAAG,MAAM;aAE~~/C,~~aAAa,UAAU,MAAM,KAAG,MAAM;aAEtC,QAAQ,UAAU,MAAM,aAAa,MAAM,MAAM,MAAM,KAAG,MAAM;aAEhE,UAAU,UAAU,MAAM,KAAG,MAAM;CAE3B,~~CAAC;AAEX;;;;;GAKG;AACH,eAAO,MAAM,qBAAqB,MAAM,CAAC;AAoBzC,eAAO,MAAM,oCAAoC,OAAQ,CAAC;AAC1D,eAAO,MAAM,kCAAkC,EAAE,MAA8B,CAAC;AAChF,eAAO,MAAM,uCAAuC,aAAgB,CAAC;AAErE,mDAAmD;AACnD,MAAM,WAAW,qBAAqB;IACpC,sEAAsE;IACtE,MAAM,EAAE,SAAS,GAAG;QAAE,UAAU,EAAE,MAAM,CAAA;KAAE,CAAC;IAC3C,OAAO,EACH,KAAK,GACL;QAAE,WAAW,EAAE,MAAM,CAAC;QAAC,QAAQ,EAAE,MAAM,CAAC;QAAC,aAAa,EAAE,MAAM,CAAA;KAAE,CAAC;CACtE;AAED,wEAAwE;AACxE,wBAAgB,oBAAoB,CAClC,MAAM,EAAE,aAAa,GAAG,SAAS,GAChC,qBAAqB,CA8BvB"}
1	+ {"version":3,"file":"storage.d.ts","sourceRoot":"","sources":["../src/storage.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA2CG;AAGH,OAAO,KAAK,EAAE,cAAc,EAAE,MAAM,SAAS,CAAC;AAC9C,OAAO,KAAK,EAAE,eAAe,EAAE,MAAM,YAAY,CAAC;AAClD,OAAO,KAAK,EAAE,UAAU,EAAE,MAAM,2BAA2B,CAAC;AAE5D,OAAO,KAAK,EAAE,SAAS,EAAE,MAAM,YAAY,CAAC;AAM5C,wEAAwE;AACxE,MAAM,WAAW,YAAY;IAC3B,GAAG,EAAE,MAAM,CAAC;IACZ,KAAK,EAAE,SAAS,CAAC;CAClB;AAED,yCAAyC;AACzC,MAAM,WAAW,cAAc;IAC7B,uEAAuE;IACvE,SAAS,EAAE,MAAM,CAAC;IAClB,sEAAsE;IACtE,WAAW,EAAE,MAAM,CAAC;IACpB;;;;;OAKG;IACH,MAAM,EAAE,QAAQ,GAAG,UAAU,GAAG,SAAS,CAAC;CAC3C;AAED,MAAM,WAAW,aAAa;IAC5B;;;;;;OAMG;IACH,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB;;;;;;;;;OASG;IACH,OAAO,CAAC,EAAE,oBAAoB,GAAG,KAAK,CAAC;CACxC;AAED,mFAAmF;AACnF,MAAM,WAAW,oBAAoB;IACnC,wFAAwF;IACxF,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB;;;;OAIG;IACH,QAAQ,CAAC,EAAE,MAAM,CAAC;IAClB;;;;;OAKG;IACH,aAAa,CAAC,EAAE,MAAM,CAAC;CACxB;AAED;;;;;GAKG;AACH,MAAM,WAAW,iBAAiB;IAChC,mEAAmE;IACnE,GAAG,CAAC,GAAG,EAAE,eAAe,EAAE,GAAG,EAAE,cAAc,GAAG,IAAI,GAAG,OAAO,CAAC,IAAI,CAAC,CAAC;IACrE,gEAAgE;IAChE,MAAM,CAAC,KAAK,EAAE,MAAM,EAAE,GAAG,EAAE,cAAc,GAAG,IAAI,GAAG,OAAO,CAAC,IAAI,CAAC,CAAC;IACjE,+EAA+E;IAC/E,IAAI,CAAC,GAAG,EAAE,cAAc,GAAG,eAAe,EAAE,GAAG,OAAO,CAAC,eAAe,EAAE,CAAC,CAAC;CAC3E;AAED;;;;;;GAMG;AACH,MAAM,WAAW,cAAc;IAC7B,gEAAgE;IAChE,SAAS,CAAC,MAAM,EAAE,cAAc,EAAE,GAAG,EAAE,cAAc,GAAG,IAAI,GAAG,OAAO,CAAC,IAAI,CAAC,CAAC;IAC7E,4EAA4E;IAC5E,WAAW,CAAC,CAAC,QAAQ,EAAE,UAAU,EAAE,GAAG,EAAE,cAAc,GAAG,IAAI,GAAG,OAAO,CAAC,IAAI,CAAC,CAAC;IAC9E,6EAA6E;IAC7E,WAAW,CAAC,CACV,GAAG,EAAE,cAAc,GAClB,UAAU,GAAG,SAAS,GAAG,IAAI,GAAG,OAAO,CAAC,UAAU,GAAG,SAAS,GAAG,IAAI,CAAC,CAAC;CAC3E;AAED,MAAM,WAAW,aAAa;IAC5B,6DAA6D;IAC7D,IAAI,CAAC,EAAE,MAAM,CAAC;IACd,+CAA+C;IAC/C,MAAM,CAAC,EAAE,aAAa,CAAC;IACvB;;;OAGG;IACH,KAAK,CAAC,EAAE,iBAAiB,CAAC;IAC1B;;;OAGG;IACH,GAAG,CAAC,EAAE,cAAc,CAAC;IACrB;;;;;;OAMG;IACH,GAAG,CAAC,GAAG,EAAE,MAAM,EAAE,KAAK,EAAE,SAAS,EAAE,GAAG,EAAE,cAAc,GAAG,IAAI,GAAG,OAAO,CAAC,IAAI,CAAC,CAAC;IAC9E,kEAAkE;IAClE,MAAM,CAAC,CAAC,GAAG,EAAE,MAAM,EAAE,GAAG,EAAE,cAAc,GAAG,IAAI,GAAG,OAAO,CAAC,IAAI,CAAC,CAAC;IAChE;;;;;;;OAOG;IACH,GAAG,CAAC,CACF,GAAG,EAAE,MAAM,EACX,GAAG,EAAE,cAAc,GAClB,SAAS,GAAG,SAAS,GAAG,IAAI,GAAG,OAAO,CAAC,SAAS,GAAG,SAAS,GAAG,IAAI,CAAC,CAAC;IACxE;;;;OAIG;IACH,IAAI,CAAC,CACH,MAAM,EAAE,MAAM,EACd,GAAG,EAAE,cAAc,GAClB,YAAY,EAAE,GAAG,OAAO,CAAC,YAAY,EAAE,CAAC,CAAC;CAC7C;AAED,MAAM,MAAM,iBAAiB,GAAG,aAAa,GAAG;IAC9C,QAAQ,CAAC,YAAY,EAAE,SAAS,CAAC;CAClC,CAAC;AAEF;;;GAGG;AACH,wBAAgB,aAAa,CAAC,MAAM,EAAE,aAAa,GAAG,iBAAiB,CAuCtE;AAMD;;;GAGG;AACH,eAAO,MAAM,gBAAgB,EAAG,aAAsB,CAAC;AAEvD;;;;;;;;;;;;;;;;;;;;;GAqBG;AACH,eAAO,MAAM,WAAW;aACtB,OAAO,UAAU,MAAM,aAAa,MAAM,KAAG,MAAM;aAEnD,aAAa,UAAU,MAAM,KAAG,MAAM;aAEtC,aAAa,UACJ,MAAM,aACF,MAAM,cACL,MAAM,KACjB,MAAM;aAET,mBAAmB,UAAU,MAAM,aAAa,MAAM,KAAG,MAAM;aAE/D,YAAY,UACH,MAAM,aACF,MAAM,mBACA,MAAM,KACtB,MAAM;aAET,QAAQ,UAAU,MAAM,cAAc,MAAM,KAAG,MAAM;aAErD,cAAc,UAAU,MAAM,KAAG,MAAM;CAE/B,CAAC;AAEX;;;;;GAKG;AACH,eAAO,MAAM,qBAAqB,MAAM,CAAC;AAoBzC,eAAO,MAAM,oCAAoC,OAAQ,CAAC;AAC1D,eAAO,MAAM,kCAAkC,EAAE,MAA8B,CAAC;AAChF,eAAO,MAAM,uCAAuC,aAAgB,CAAC;AAErE,mDAAmD;AACnD,MAAM,WAAW,qBAAqB;IACpC,sEAAsE;IACtE,MAAM,EAAE,SAAS,GAAG;QAAE,UAAU,EAAE,MAAM,CAAA;KAAE,CAAC;IAC3C,OAAO,EACH,KAAK,GACL;QAAE,WAAW,EAAE,MAAM,CAAC;QAAC,QAAQ,EAAE,MAAM,CAAC;QAAC,aAAa,EAAE,MAAM,CAAA;KAAE,CAAC;CACtE;AAED,wEAAwE;AACxE,wBAAgB,oBAAoB,CAClC,MAAM,EAAE,aAAa,GAAG,SAAS,GAChC,qBAAqB,CA8BvB"}

package/dist/storage.js CHANGED Viewed

@@ -4,10 +4,11 @@
  * Author `agent/storage.ts` with {@link defineStorage} to mirror the
  * framework's durable state into storage you own (a database, S3, a data
  * pipeline, …). Without it, state lives under `--state-root` on local disk
- * only (and eval/A/B history follows the narrower `persistRuns` /
- * `persistSamples` / `persistSnapshots` hooks).
+ * only, and eval/A/B history stays in process memory.
  *
- * The sink is a plain key-value store — four functions, no schema:
+ * The sink is a plain key-value store — four functions, no schema — plus
+ * two optional dedicated tables ({@link StorageConfig.evals} and
+ * {@link StorageConfig.abs}) for eval-run and A/B history:
  *
  * ```ts
  * import { defineStorage } from "@anysphere/agent-serve/storage";
@@ -17,6 +18,8 @@
  *   get: (key) => db.get(key),
  *   delete: (key) => db.delete(key),
  *   list: (prefix) => db.listByPrefix(prefix), // [{ key, value }] in key order
+ *   evals: { put: ..., delete: ..., list: ... }, // eval-runs table
+ *   abs: { putSample: ... },                     // A/B table
  * });
  * ```
  *
@@ -55,6 +58,24 @@ export function defineStorage(config) {
             throw new Error(`defineStorage: config.${hook} must be a function when set`);
         }
     }
+    if (config.evals !== undefined) {
+        for (const hook of ["put", "delete", "list"]) {
+            if (typeof config.evals[hook] !== "function") {
+                throw new Error(`defineStorage: evals.${hook} must be a function when the evals table is set`);
+            }
+        }
+    }
+    if (config.abs !== undefined) {
+        if (typeof config.abs.putSample !== "function") {
+            throw new Error("defineStorage: abs.putSample must be a function when the abs table is set");
+        }
+        for (const hook of ["putSnapshot", "getSnapshot"]) {
+            const value = config.abs[hook];
+            if (value !== undefined && typeof value !== "function") {
+                throw new Error(`defineStorage: abs.${hook} must be a function when set`);
+            }
+        }
+    }
     // Validate eagerly so a bad policy fails at discovery, not first flush.
     resolveStoragePolicy(config.policy);
     return brandDefinition("storage", config);
@@ -85,9 +106,9 @@ export const STORAGE_KEY_ROOT = "agentkit/v1";
  * | `agentkit/v1/{agent}/session-events/{sessionId}/{index}` | `SessionEvent[]` chunk (index = first event's index, zero-padded) |
  * | `agentkit/v1/{agent}/continuation/{channelId}/{token}` | `{ sessionId }` |
  * | `agentkit/v1/{agent}/reminder/{reminderId}` | `ReminderRecord` |
- * | `agentkit/v1/{agent}/eval-run/{runId}` | `EvalRunSnapshot` |
- * | `agentkit/v1/{agent}/ab-sample/{sessionId}/{at}` | `ABMetricSample` |
- * | `agentkit/v1/{agent}/ab-snapshot` | latest aggregate `ABSnapshot` |
+ *
+ * Eval-run and A/B history do not flow through this KV scheme — they have
+ * dedicated tables ({@link StorageConfig.evals} / {@link StorageConfig.abs}).
  */
 export const storageKeys = {
     session: (agent, sessionId) => `${STORAGE_KEY_ROOT}/${agent}/session/${sessionId}`,
@@ -97,10 +118,6 @@ export const storageKeys = {
     continuation: (agent, channelId, continuationKey) => `${STORAGE_KEY_ROOT}/${agent}/continuation/${keySegment(channelId)}/${keySegment(continuationKey)}`,
     reminder: (agent, reminderId) => `${STORAGE_KEY_ROOT}/${agent}/reminder/${reminderId}`,
     reminderPrefix: (agent) => `${STORAGE_KEY_ROOT}/${agent}/reminder/`,
-    evalRun: (agent, runId) => `${STORAGE_KEY_ROOT}/${agent}/eval-run/${runId}`,
-    evalRunPrefix: (agent) => `${STORAGE_KEY_ROOT}/${agent}/eval-run/`,
-    abSample: (agent, sessionId, at) => `${STORAGE_KEY_ROOT}/${agent}/ab-sample/${sessionId}/${at}`,
-    abSnapshot: (agent) => `${STORAGE_KEY_ROOT}/${agent}/ab-snapshot`,
 };
 /**
  * Max bytes a caller-controlled key segment may occupy after URI-encoding.

package/docs/README.md CHANGED Viewed

@@ -47,11 +47,12 @@ Pick your entry point based on your goal.
 | --- | --- |
 | New to agentkit | [Quickstart](./quickstart.md) (PR approver), then [Concepts](./concepts.md) |
 | Building a new agent with Cursor | [Scaffold an agent with Cursor](./scaffolding-agents.md) |
+| Learning from working agents | [Example agents](./example-agents/index.md) |
 | Wiring an agent to Slack | [Slack guide](./guides/slack.md) |
 | Wiring an agent to GitHub webhooks | [GitHub guide](./guides/github.md) |
 | Making an existing agent measurably better | [Evals](./evals.md), then [Hillclimbing](./hillclimbing.md) |
 | Comparing variants on live traffic | [Live A/B metrics](./ab.md) |
-| Deploying an agent to a server | [Deployment](./deployment.md) |
+| Deploying with Cursor or on your own infrastructure | [Deployment](./deployment.md) |
 | Debugging something that misbehaves | [Fix common agent problems](./troubleshooting.md) |
 ## The documentation
@@ -89,10 +90,37 @@ Pick your entry point based on your goal.
 - [Cloud runtime](./guides/cloud-runtime.md): run turns on Cursor cloud
   agents instead of the local harness.
+**Example agents**
+- [Choose the right example](./example-agents/index.md): compare all eleven
+  agents by runtime, channels, tools, state, and architecture.
+- [Weather agent](./example-agents/weather-agent.md): explore tools, MCP,
+  approvals, skills, subagents, schedules, hooks, A/B metrics, and evals.
+- [Slack agent](./example-agents/slack-agent.md): put a minimal agent in
+  Slack through an account-linked transport.
+- [Concierge](./example-agents/concierge.md): delegate work to a peer agent
+  with its own context and sessions.
+- [Playbook router](./example-agents/benny.md): route Slack intake through inherited
+  repository playbooks.
+- [PR evidence reviewer](./example-agents/bugbot.md): review a host-prepared,
+  diff-first pull-request evidence tree.
+- [Approval Buddy](./example-agents/approval-buddy.md): keep approval policy
+  in code while subagents supply review findings.
+- [Security Reviewer](./example-agents/security-reviewer.md): run a staged,
+  parallel security pipeline with live playground progress.
+- [Remote PR coordinator](./example-agents/fsd.md): hand PR triage from local
+  chat and webhooks to durable remote sessions.
+- [Knowledge base](./example-agents/knowledge-base.md): turn conversations
+  about people, systems, decisions, and preferences into shared markdown.
+- [Codebase wiki](./example-agents/codebase-wiki.md): ingest merged PRs into
+  per-feature pages with a daily digest schedule.
+- [Codeowners review](./example-agents/codeowners-review.md): route PR
+  reviews by ownership to per-area playbooks and aggregate verdicts.
 **Operating**
-- [Deployment](./deployment.md): auth, state, credentials, Docker, and
-  running on a server.
+- [Deployment](./deployment.md): Cursor-managed hosting, self-hosting,
+  auth, state, and operations.
 - [Fix common agent problems](./troubleshooting.md): symptom to cause,
   in plain language.
@@ -132,8 +160,9 @@ comes from depends on where you run.
 > [!NOTE]
 > The framework is being renamed from agent-serve to agentkit,
-> and these docs use the new names throughout. Until the code rename
-> ships, substitute the current identifiers where they differ:
+> and CLI examples use the new `agentkit` name. Paths, package imports,
+> and environment variables keep their current names until the code rename
+> ships:
 >
 > | Docs say | Current name |
 > | --- | --- |