@noodleseed/agent-kit 0.19.0 → 0.19.1

package/manifest.json

@@ -1,9 +1,9 @@
 {
-  "packageVersion": "0.19.0",
+  "packageVersion": "0.19.1",
   "files": [
     {
       "path": "skills/codex/SKILL.md",
-      "sha256": "590f306f725fbc64e576fbcba6dad4bb4b163bb2cbfc799cad04adac7645a77f",
+      "sha256": "076f93b1c9e1e2ac9b5efe4ae2d80e4aed9ecb75e0da9257f327085360d802f1",
       "agentTarget": "codex"
     },
     {
@@ -28,7 +28,7 @@
     },
     {
       "path": "skills/codex/references/authoring-workflow.md",
-      "sha256": "b078cfc3481824ed09a7445346a793a89387b0a07de500f7bb5343487ec62e58",
+      "sha256": "fd2a67d3ba67f640edddca6d30ed8d38f278f0f3e9ab1c1669ace2aef8797c01",
       "agentTarget": "codex"
     },
     {
@@ -193,7 +193,7 @@
     },
     {
       "path": "skills/codex/examples/acme-tasks/README.md",
-      "sha256": "f97b264bd7c696ffd8ddbba846cdcd8c2e1b0940738e94d057654b5e42d14d1a",
+      "sha256": "1d65cde8d1b47ade9e895ef99a92a0aec9d9ffe0c578f0d2dc4bafa0bcccca26",
       "agentTarget": "codex"
     },
     {
@@ -348,7 +348,7 @@
     },
     {
       "path": "skills/claude-code/SKILL.md",
-      "sha256": "09ceee39251194a1528986cc7df57970218e865646779ce01e98872ccadfaffd",
+      "sha256": "7b487e019e82c28ec2ef1f5ffcb54f8ea15e3b62ecaf10a9de7375a692a4d1f7",
       "agentTarget": "claude-code"
     },
     {
@@ -373,7 +373,7 @@
     },
     {
       "path": "skills/claude-code/references/authoring-workflow.md",
-      "sha256": "b078cfc3481824ed09a7445346a793a89387b0a07de500f7bb5343487ec62e58",
+      "sha256": "fd2a67d3ba67f640edddca6d30ed8d38f278f0f3e9ab1c1669ace2aef8797c01",
       "agentTarget": "claude-code"
     },
     {
@@ -538,7 +538,7 @@
     },
     {
       "path": "skills/claude-code/examples/acme-tasks/README.md",
-      "sha256": "f97b264bd7c696ffd8ddbba846cdcd8c2e1b0940738e94d057654b5e42d14d1a",
+      "sha256": "1d65cde8d1b47ade9e895ef99a92a0aec9d9ffe0c578f0d2dc4bafa0bcccca26",
       "agentTarget": "claude-code"
     },
     {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@noodleseed/agent-kit",
-  "version": "0.19.0",
+  "version": "0.19.1",
   "private": false,
   "description": "Self-checking, self-updating agent skills for the Noodle Seed CLI. Authored in this repo by @noodle-borg/agent-kit; this is the published, independently-versioned canonical skills artifact the CLI fetches and verifies.",
   "license": "Apache-2.0",

package/skills/claude-code/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: noodle-seed
 description: Use when building, validating, testing, deploying, or operating a local or hosted Noodle Seed MCP server or app authored in TypeScript with the noodle CLI.
-version: 0.19.0
+version: 0.19.1
 hash: b4952708b1c38f55
 ---
 

package/skills/claude-code/examples/acme-tasks/README.md

@@ -102,5 +102,8 @@ uses the Web Component or React wrapper and never receives the embed client or m
 `noodle check --target embedded-assistant`, then create the backend credential with
 `noodle assistant clients create` after deployment.
 
+Install the independently versioned embed SDK in the customer web application with
+`pnpm add @noodleseed/assistant`.
+
 This example has no connector secrets and does not include tokens, caller-key mechanisms, or
 `.env.noodle` values. All tasks are fictional seed data.

package/skills/claude-code/references/authoring-workflow.md

@@ -200,7 +200,7 @@ Author managed config as `secret("NAME")` / `variable("NAME")` and operate it wi
 
 ## Embedded assistant
 
-To place the same server tools inside the SaaS web app, declare the shared identity once with `branding: { name: "Acme", accent: "#3157D5", surface: "#FFFFFF", surfaceDark: "#161B26", theme: { dark: { accent: "#9FB4FF" } } }`, then add `assistant: embeddedAssistant({ model: openAICompatible({ baseUrl: variable("ASSISTANT_MODEL_BASE_URL"), model: variable("ASSISTANT_MODEL"), apiKey: secret("ASSISTANT_MODEL_API_KEY") }), allowedOrigins: ["https://app.example.com"], layout: { mode: "floating", position: "bottom-right" } })`. Widgets and the embed inherit the same brand kit. Keep origins exact. The customer backend uses `@noodleseed/assistant/server` to exchange its already-authenticated user for a short-lived browser session; client and model secrets never enter the browser. Run `noodle check --target embedded-assistant` before deploy, then manage backend credentials with `noodle assistant clients create|list|rotate|revoke`.
+To place the same server tools inside the SaaS web app, declare the shared identity once with `branding: { name: "Acme", accent: "#3157D5", surface: "#FFFFFF", surfaceDark: "#161B26", theme: { dark: { accent: "#9FB4FF" } } }`, then add `assistant: embeddedAssistant({ model: openAICompatible({ baseUrl: variable("ASSISTANT_MODEL_BASE_URL"), model: variable("ASSISTANT_MODEL"), apiKey: secret("ASSISTANT_MODEL_API_KEY") }), allowedOrigins: ["https://app.example.com"], layout: { mode: "floating", position: "bottom-right" } })`. Widgets and the embed inherit the same brand kit. Keep origins exact. Install the independently versioned embed SDK with `pnpm add @noodleseed/assistant`; the customer backend uses its `/server` export to exchange the already-authenticated user for a short-lived browser session, while the browser uses the package root or `/react` export. Client and model secrets never enter the browser. Run `noodle check --target embedded-assistant` before deploy, then manage backend credentials with `noodle assistant clients create|list|rotate|revoke`.
 
 ## Boundaries
 

package/skills/codex/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: noodle-seed
 description: Use when building, validating, testing, deploying, or operating a local or hosted Noodle Seed MCP server or app authored in TypeScript with the noodle CLI.
-version: 0.19.0
+version: 0.19.1
 hash: 39c6a606c728d164
 ---
 

package/skills/codex/examples/acme-tasks/README.md

@@ -102,5 +102,8 @@ uses the Web Component or React wrapper and never receives the embed client or m
 `noodle check --target embedded-assistant`, then create the backend credential with
 `noodle assistant clients create` after deployment.
 
+Install the independently versioned embed SDK in the customer web application with
+`pnpm add @noodleseed/assistant`.
+
 This example has no connector secrets and does not include tokens, caller-key mechanisms, or
 `.env.noodle` values. All tasks are fictional seed data.

package/skills/codex/references/authoring-workflow.md

@@ -200,7 +200,7 @@ Author managed config as `secret("NAME")` / `variable("NAME")` and operate it wi
 
 ## Embedded assistant
 
-To place the same server tools inside the SaaS web app, declare the shared identity once with `branding: { name: "Acme", accent: "#3157D5", surface: "#FFFFFF", surfaceDark: "#161B26", theme: { dark: { accent: "#9FB4FF" } } }`, then add `assistant: embeddedAssistant({ model: openAICompatible({ baseUrl: variable("ASSISTANT_MODEL_BASE_URL"), model: variable("ASSISTANT_MODEL"), apiKey: secret("ASSISTANT_MODEL_API_KEY") }), allowedOrigins: ["https://app.example.com"], layout: { mode: "floating", position: "bottom-right" } })`. Widgets and the embed inherit the same brand kit. Keep origins exact. The customer backend uses `@noodleseed/assistant/server` to exchange its already-authenticated user for a short-lived browser session; client and model secrets never enter the browser. Run `noodle check --target embedded-assistant` before deploy, then manage backend credentials with `noodle assistant clients create|list|rotate|revoke`.
+To place the same server tools inside the SaaS web app, declare the shared identity once with `branding: { name: "Acme", accent: "#3157D5", surface: "#FFFFFF", surfaceDark: "#161B26", theme: { dark: { accent: "#9FB4FF" } } }`, then add `assistant: embeddedAssistant({ model: openAICompatible({ baseUrl: variable("ASSISTANT_MODEL_BASE_URL"), model: variable("ASSISTANT_MODEL"), apiKey: secret("ASSISTANT_MODEL_API_KEY") }), allowedOrigins: ["https://app.example.com"], layout: { mode: "floating", position: "bottom-right" } })`. Widgets and the embed inherit the same brand kit. Keep origins exact. Install the independently versioned embed SDK with `pnpm add @noodleseed/assistant`; the customer backend uses its `/server` export to exchange the already-authenticated user for a short-lived browser session, while the browser uses the package root or `/react` export. Client and model secrets never enter the browser. Run `noodle check --target embedded-assistant` before deploy, then manage backend credentials with `noodle assistant clients create|list|rotate|revoke`.
 
 ## Boundaries