@adobe/design-data-agent-mcp 1.12.0 → 1.13.0

package/README.md

@@ -1,6 +1,6 @@
 # `@adobe/design-data-agent-mcp`
 
-MCP server and Claude Code skill for the [Spectrum Design Data](../../packages/design-data/) agent surface. Read tools (`primer`, `resolve_token`, `query_tokens`, `describe_component`, `describe_guideline`, `list_guidelines`) run fully in-process via `@adobe/design-data-wasm` — no CLI binary required for those. Only `authoring_session_step_intent` still invokes the native binary (for NLP suggest ranking, not yet on the wasm surface).
+MCP server and Claude Code skill for the [Spectrum Design Data](../../packages/design-data/) agent surface. Read tools (`primer`, `resolve_token`, `query_tokens`, `suggest_token`, `describe_component`, `describe_guideline`, `list_guidelines`) run fully in-process via `@adobe/design-data-wasm` — no CLI binary required for those. `suggest_token` uses the wasm `suggest` API directly; `authoring_session_step_intent` still invokes the native binary for session-state-aware ranking within an authoring flow.
 
 ## Install
 
@@ -152,6 +152,7 @@ node tools/design-data-agent-mcp/src/index.js
 | `primer`             | Load full token taxonomy, component list, and field definitions |
 | `resolve_token`      | Resolve a token property to its literal value                   |
 | `query_tokens`       | Filter tokens by expression                                     |
+| `suggest_token`      | Suggest tokens matching a natural-language intent               |
 | `describe_component` | Fetch component schema and token bindings                       |
 | `describe_guideline` | Fetch guideline documentBlocks content and metadata             |
 | `list_guidelines`    | List guideline slugs and metadata, optionally by category       |

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@adobe/design-data-agent-mcp",
-  "version": "1.12.0",
+  "version": "1.13.0",
   "description": "MCP server and Claude Code skill for the design-data agent surface — read tools run in-process via wasm",
   "type": "module",
   "main": "./src/index.js",

package/skills/design-data/SKILL.md

@@ -6,12 +6,12 @@ description: >
   system, token lookup, spec-conformance, drift detection, or token authoring on custom data.
 metadata:
   author: adobe
-  version: "1.12.0"
+  version: "1.13.0"
 when_to_use: >
   Trigger on: design system, design tokens, spec-conformant, drift, validate tokens, token
   authoring, custom dataset, DESIGN_DATA_PATH, design-data validate, design-data diff,
   design-data write, product-context.json.
-allowed-tools: mcp__design-data-agent__primer, mcp__design-data-agent__query_tokens, mcp__design-data-agent__resolve_token, mcp__design-data-agent__describe_component, mcp__design-data-agent__validate_usage, mcp__design-data-agent__diff_datasets, mcp__design-data-agent__write, mcp__design-data-agent__start_authoring_session, mcp__design-data-agent__authoring_session_step_intent, mcp__design-data-agent__authoring_session_step_classification, mcp__design-data-agent__authoring_session_step_values, mcp__design-data-agent__authoring_session_commit, mcp__design-data-agent__authoring_session_cancel, mcp__design-data-agent__authoring_session_get, mcp__design-data-agent__authoring_session_list
+allowed-tools: mcp__design-data-agent__primer, mcp__design-data-agent__query_tokens, mcp__design-data-agent__suggest_token, mcp__design-data-agent__resolve_token, mcp__design-data-agent__describe_component, mcp__design-data-agent__validate_usage, mcp__design-data-agent__diff_datasets, mcp__design-data-agent__write, mcp__design-data-agent__start_authoring_session, mcp__design-data-agent__authoring_session_step_intent, mcp__design-data-agent__authoring_session_step_classification, mcp__design-data-agent__authoring_session_step_values, mcp__design-data-agent__authoring_session_commit, mcp__design-data-agent__authoring_session_cancel, mcp__design-data-agent__authoring_session_get, mcp__design-data-agent__authoring_session_list
 ---
 
 # design-data agent skill
@@ -102,6 +102,15 @@ $schema=https://spectrum.adobe.com/page/design-token/
 
 > **Exit codes:** `0` = matches found; empty array = no matches (not an error).
 
+### Suggest a token from a description — `suggest_token`
+
+Required: `intent` (string) — a natural-language description of the design need,
+such as `"primary CTA button background color"`.
+
+Optional: `limit` (number, default `5`) — the maximum number of suggestions to return.
+Results are ranked by confidence using token names, name-object fields, and description
+text. Use this when the user describes what they need rather than knowing a token name.
+
 ***
 
 ## Component info — `describe_component`
@@ -163,8 +172,9 @@ Helper tools: `authoring_session_get` (inspect state), `authoring_session_list`
 `authoring_session_commit` accepts an optional `schema_path` to override the schemas directory
 for Layer-1 JSON-Schema validation before writing.
 
-> **Note:** `authoring_session_step_intent` (NLP suggestion ranking) still delegates to the
-> `design-data` CLI because the NLP `suggest` API is not yet on the wasm surface.
+> **Note:** The standalone `suggest_token` tool calls the wasm `suggest` API directly.
+> `authoring_session_step_intent` still delegates to the `design-data` CLI for
+> session-state-aware ranking within an authoring flow.
 
 ***
 

package/src/tools/read.js

@@ -20,8 +20,9 @@
  * @adobe/design-data-mcp package's `design-data-guideline` tool via the
  * `@adobe/design-data/guideline` module, rather than duplicated here.
  *
- * Note: authoring_session_step_intent in authoring.js still uses the CLI because
- * the NLP suggest ranking is not yet on the wasm surface.
+ * suggest_token uses the wasm suggest surface directly. The separate
+ * authoring_session_step_intent flow still uses the CLI for session-state-aware
+ * ranking within an authoring session.
  *
  * Cascade scope (see cascade-bootstrap.js / spectrum-design-data-h890.14): once a
  * `.design-data.toml` cascade is resolved, primer/resolve_token/query_tokens/
@@ -286,6 +287,36 @@ export function createReadTools() {
       },
     },
 
+    {
+      name: "suggest_token",
+      description:
+        "Suggest Spectrum tokens matching a natural-language intent using Jaccard similarity " +
+        "scoring over token name segments, name-object fields, and description text. " +
+        "Returns matches ranked by confidence with token name, layer, value, and name object. " +
+        "Use when the user describes what they need rather than knowing the token name.",
+      inputSchema: {
+        type: "object",
+        required: ["intent"],
+        properties: {
+          intent: {
+            type: "string",
+            description:
+              'Natural-language description of the design need, e.g. "primary CTA button background color"',
+          },
+          limit: {
+            type: "number",
+            description: "Maximum number of suggestions to return (default: 5)",
+            default: 5,
+          },
+        },
+        additionalProperties: false,
+      },
+      async handler({ intent, limit = 5 }) {
+        const ds = await getDataset();
+        return ds.suggest(intent, undefined, limit);
+      },
+    },
+
     {
       name: "describe_component",
       description: