@coalex-ai/sdk 2.2.0 → 2.2.1-beta.670

package/README.md

@@ -28,7 +28,7 @@ await autoInstrumentAsync();
 
 // 3. Wrap agent invocations for tracing
 // `version` is the build this agent is running. It is optional today and omitting it logs a
-// warning (COA-2636): a trace with no version cannot be attributed to a build, and the
+// warning once per agent: a trace with no version cannot be attributed to a build, and the
 // dashboard records it as `unknown`. It will become required in a future major release.
 const result = await coalexContext({ agentId: "support-agent", version: "1.4.0" }, async () => {
   // Your AI agent logic here
@@ -52,6 +52,28 @@ register({
 });
 ```
 
+To have the agent appear in the Console without anyone registering it by hand, pass `declare`. The
+declaration is best-effort: a failure is logged and the agent still starts and traces. Without
+`declare`, `register()` makes no network call.
+
+```typescript
+register({
+  apiKey: "your-api-key",
+  serviceName: "my-agent",
+  declare: {
+    endpoint: "https://my-agent.example.com/turn", // where the Console POSTs a turn
+    tools: ["wire_funds"],                         // tool names only
+    resumeUrl: "https://my-agent.example.com/resume", // if the agent supports being paused
+    displayName: "Payments agent",                 // title of a new chat in the Console
+    description: "Ask me about wire transfers.",   // shown as the agent's first message
+    suggestedPrompts: ["Show pending wires"],      // shown as buttons that send that text
+  },
+});
+```
+
+An agent that declares no `displayName`, `description` or `suggestedPrompts` is shown by its name
+alone. Whether a tool is governed and who may approve it are set in the dashboard, not here.
+
 ### `autoInstrumentAsync(options?)`
 
 Instrument AI libraries using dynamic `import()`. Use this in ESM projects.
@@ -131,14 +153,20 @@ Submit an AI output for evaluation. Returns the governance decision.
 
 ```typescript
 const decision = await evaluate({
-  traceId: "abc-123",
+  requestId: "abc-123",
   input: { question: "What is the policy?" },
   output: { answer: "The policy states..." },
   metrics: { answer: ["f1", "rouge_l"] },
+  toolName: "wire_funds", // optional: the tool this call is about
 });
-// { status: "auto_approved" | "escalated" | "rejected", riskScore: 0.12, escalationId?: "..." }
+// { status: "auto_approved" | "escalated", riskScore: 0.12, escalationId?: "..." }
 ```
 
+`toolName` is stored on the escalation and in the decision's audit-chain entry, so an auditor reads
+which action a person approved, declined or let expire. Only the name is sent there; the tool's
+arguments stay in `input` and `output`. Omitted, the platform reads a `tool` key from `output`,
+`input` or `metadata`. `traceId` is a deprecated alias of `requestId`.
+
 Available metrics: `f1`, `word_overlap`, `bleu`, `rouge_l`, `exact_match`, `levenshtein`, `contains`.
 
 ### `resolve(options)`

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@coalex-ai/sdk",
-  "version": "2.2.0",
+  "version": "2.2.1-beta.670",
   "description": "Coalex TypeScript SDK for AI governance observability",
   "license": "Apache-2.0",
   "repository": {
@@ -46,7 +46,7 @@
     "vitest": "^5.0.2",
     "eslint": "^10.0.0",
     "@eslint/js": "^10.0.0",
-    "typescript-eslint": "^8.70.1"
+    "typescript-eslint": "^8.71.0"
   },
   "peerDependencies": {
     "@opentelemetry/instrumentation": ">=0.200.0"