@pauly4010/evalai-sdk 1.4.1 → 1.5.5

package/README.md

@@ -1,641 +1,303 @@
 # @pauly4010/evalai-sdk
 
-[![npm version](https://img.shields.io/npm/v/@pauly4010/evalai-sdk.svg)](https://www.npmjs.com/package/@pauly4010/evalai-sdk)
+[![npm version](https://img.shields.io/npm/v/@pauly4010/evalai-sdk.svg)](https://www.npmjs.com/package/@pauly4010/evalai-sdk)  
 [![npm downloads](https://img.shields.io/npm/dm/@pauly4010/evalai-sdk.svg)](https://www.npmjs.com/package/@pauly4010/evalai-sdk)
 
-Official TypeScript/JavaScript SDK for the AI Evaluation Platform. Build confidence in your AI systems with comprehensive evaluation tools.
+**Stop LLM regressions in CI in minutes.**
 
-## Installation
+Evaluate locally in 60 seconds. Add an optional CI gate in 2 minutes.  
+No lock-in — remove by deleting `evalai.config.json`.
+
+---
+
+# 🚀 1) 60 seconds: Run locally (no account)
+
+Install, run, get a score.  
+No EvalAI account. No API key. No dashboard required.
 
 ```bash
-npm install @pauly4010/evalai-sdk
-# or
-yarn add @pauly4010/evalai-sdk
-# or
-pnpm add @pauly4010/evalai-sdk
-```
+npm install @pauly4010/evalai-sdk openai
+import { openAIChatEval } from "@pauly4010/evalai-sdk";
+
+await openAIChatEval({
+  name: "chat-regression",
+  cases: [
+    { input: "Hello", expectedOutput: "greeting" },
+    { input: "2 + 2 = ?", expectedOutput: "4" },
+  ],
+});
+Set:
+
+OPENAI_API_KEY=...
+✅ Vitest integration (recommended)
+import {
+  openAIChatEval,
+  extendExpectWithToPassGate,
+} from "@pauly4010/evalai-sdk";
+import { expect } from "vitest";
+
+extendExpectWithToPassGate(expect);
+
+it("passes gate", async () => {
+  const result = await openAIChatEval({
+    name: "chat-regression",
+    cases: [
+      { input: "Hello", expectedOutput: "greeting" },
+      { input: "2 + 2 = ?", expectedOutput: "4" },
+    ],
+  });
 
-## Environment Support
+  expect(result).toPassGate();
+});
+Example output
+PASS 2/2 (score: 100)
 
-This SDK works in both **Node.js** and **browsers**, with some features having specific requirements:
+Tip: Want dashboards and history?
+Set EVALAI_API_KEY and connect this to the platform.
+Failures show:
 
-### ✅ Works Everywhere (Node.js + Browser)
+FAIL 9/10 (score: 90)
+with failed cases and CI guidance.
 
-- Traces API
-- Evaluations API
-- LLM Judge API
-- Annotations API
-- Developer API (API Keys, Webhooks, Usage)
-- Organizations API
-- Assertions Library
-- Test Suites
-- Error Handling
+⚡ 2) Optional: Add a CI gate (2 minutes)
+When you're ready to gate PRs on quality and regressions:
 
-### 🟡 Node.js Only Features
+npx -y @pauly4010/evalai-sdk@^1 init
+Create an evaluation in the dashboard and paste its ID into:
 
-The following features require Node.js and **will not work in browsers**:
+{
+  "evaluationId": "42"
+}
+Add to your CI:
 
-- **Snapshot Testing** - Uses filesystem for storage
-- **Local Storage Mode** - Uses filesystem for offline development
-- **CLI Tool** - Command-line interface
-- **Export to File** - Direct file system writes
+- name: EvalAI gate
+  env:
+    EVALAI_API_KEY: ${{ secrets.EVALAI_API_KEY }}
+  run: npx -y @pauly4010/evalai-sdk@^1 check --format github --onFail import --warnDrop 1
+You’ll get:
 
-### 🔄 Context Propagation
+GitHub annotations
 
-- **Node.js**: Full async context propagation using `AsyncLocalStorage`
-- **Browser**: Basic context support (not safe across all async boundaries)
+Step summary
 
-Use appropriate features based on your environment. The SDK will throw helpful errors if you try to use Node.js-only features in a browser.
+Optional dashboard link
 
-## Quick Start
+PASS / WARN / FAIL (v1.5.5)
+EvalAI introduces a WARN band so teams can see meaningful regressions without always blocking merges.
 
-```typescript
-import { AIEvalClient } from "@pauly4010/evalai-sdk";
+Behavior
 
-// Initialize with environment variables
-const client = AIEvalClient.init();
+PASS → within thresholds
 
-// Or with explicit config
-const client = new AIEvalClient({
-  apiKey: "your-api-key",
-  organizationId: 123,
-  debug: true,
-});
-```
+WARN → regression > warnDrop but < maxDrop
 
-## Features
+FAIL → regression > maxDrop
 
-### 🎯 Evaluation Templates (v1.1.0)
+Key flags
 
-The SDK now includes comprehensive evaluation template types for different testing scenarios:
+--warnDrop → soft regression warning
 
-```typescript
-import { EvaluationTemplates } from "@pauly4010/evalai-sdk";
+--maxDrop → hard regression fail
 
-// Create evaluations with predefined templates
-await client.evaluations.create({
-  name: "Prompt Optimization Test",
-  type: EvaluationTemplates.PROMPT_OPTIMIZATION,
-  createdBy: userId,
-});
+--fail-on-flake → fail if any test is unstable
 
-// Available templates:
-// Core Testing
-EvaluationTemplates.UNIT_TESTING;
-EvaluationTemplates.OUTPUT_QUALITY;
-
-// Advanced Evaluation
-EvaluationTemplates.PROMPT_OPTIMIZATION;
-EvaluationTemplates.CHAIN_OF_THOUGHT;
-EvaluationTemplates.LONG_CONTEXT_TESTING;
-EvaluationTemplates.MODEL_STEERING;
-EvaluationTemplates.REGRESSION_TESTING;
-EvaluationTemplates.CONFIDENCE_CALIBRATION;
-
-// Safety & Compliance
-EvaluationTemplates.SAFETY_COMPLIANCE;
-
-// Domain-Specific
-EvaluationTemplates.RAG_EVALUATION;
-EvaluationTemplates.CODE_GENERATION;
-EvaluationTemplates.SUMMARIZATION;
-```
+This lets teams tune signal vs noise in CI.
 
-### 📊 Organization Resource Limits (v1.1.0)
+🔒 3) No lock-in
+To stop using EvalAI:
 
-Track your organization's resource usage and limits:
+rm evalai.config.json
+Your local openAIChatEval runs continue to work exactly the same.
 
-```typescript
-// Get current usage and limits
-const limits = await client.getOrganizationLimits();
+No account cancellation. No data export required.
 
-console.log("Traces:", {
-  usage: limits.traces_per_organization?.usage,
-  balance: limits.traces_per_organization?.balance,
-  total: limits.traces_per_organization?.included_usage,
-});
+📦 Installation
+npm install @pauly4010/evalai-sdk openai
+# or
+yarn add @pauly4010/evalai-sdk openai
+# or
+pnpm add @pauly4010/evalai-sdk openai
+🖥️ Environment Support
+This SDK works in both Node.js and browsers, with some Node-only features.
 
-console.log("Evaluations:", {
-  usage: limits.evals_per_organization?.usage,
-  balance: limits.evals_per_organization?.balance,
-  total: limits.evals_per_organization?.included_usage,
-});
+✅ Works Everywhere (Node.js + Browser)
+Traces API
 
-console.log("Annotations:", {
-  usage: limits.annotations_per_organization?.usage,
-  balance: limits.annotations_per_organization?.balance,
-  total: limits.annotations_per_organization?.included_usage,
-});
-```
+Evaluations API
 
-### 🔍 Traces
+LLM Judge API
 
-```typescript
-// Create a trace
-const trace = await client.traces.create({
-  name: "User Query",
-  traceId: "trace-123",
-  metadata: { userId: "456" },
-});
+Annotations API
 
-// List traces
-const traces = await client.traces.list({
-  limit: 10,
-  status: "success",
-});
+Developer API (API Keys, Webhooks, Usage)
 
-// Create spans
-const span = await client.traces.createSpan(trace.id, {
-  name: "LLM Call",
-  spanId: "span-456",
-  startTime: new Date().toISOString(),
-  metadata: { model: "gpt-4" },
-});
-```
+Organizations API
 
-### 📝 Evaluations
+Assertions Library
 
-```typescript
-// Create evaluation
-const evaluation = await client.evaluations.create({
-  name: "Chatbot Responses",
-  type: EvaluationTemplates.OUTPUT_QUALITY,
-  description: "Test chatbot response quality",
-  createdBy: userId,
-});
+Test Suites
 
-// Add test cases
-await client.evaluations.createTestCase(evaluation.id, {
-  input: "What is the capital of France?",
-  expectedOutput: "Paris",
-});
+Error Handling
 
-// Run evaluation
-const run = await client.evaluations.createRun(evaluation.id, {
-  status: "running",
-});
-```
+🟡 Node.js Only
+These require Node.js:
 
-### ⚖️ LLM Judge
+Snapshot Testing
 
-```typescript
-// Evaluate with LLM judge
-const result = await client.llmJudge.evaluate({
-  configId: 1,
-  input: "Translate: Hello world",
-  output: "Bonjour le monde",
-  metadata: { language: "French" },
-});
+Local Storage Mode
 
-console.log("Score:", result.result.score);
-console.log("Reasoning:", result.result.reasoning);
-```
+CLI Tool
 
-## Configuration
+Export to File
 
-### Environment Variables
+🔄 Context Propagation
+Node.js: full async context via AsyncLocalStorage
 
-```bash
-# Required
-EVALAI_API_KEY=your-api-key
+Browser: basic support (not safe across all async boundaries)
 
-# Optional
-EVALAI_ORGANIZATION_ID=123
-EVALAI_BASE_URL=https://api.example.com
-```
+🧠 AIEvalClient (Platform API)
+import { AIEvalClient } from "@pauly4010/evalai-sdk";
 
-### Client Options
+// From env
+const client = AIEvalClient.init();
 
-```typescript
-const client = new AIEvalClient({
+// Explicit
+const client2 = new AIEvalClient({
   apiKey: "your-api-key",
   organizationId: 123,
-  baseUrl: "https://api.example.com",
-  timeout: 30000,
   debug: true,
-  logLevel: "debug",
-  retry: {
-    maxAttempts: 3,
-    backoff: "exponential",
-    retryableErrors: ["RATE_LIMIT_EXCEEDED", "TIMEOUT"],
-  },
 });
-```
-
-## Error Handling
-
-```typescript
-import { EvalAIError, RateLimitError } from '@pauly4010/evalai-sdk';
+🧪 evalai CLI (v1.5.5)
+The CLI gates deployments on quality, regression, and policy.
+
+Quick start
+npx -y @pauly4010/evalai-sdk@^1 check \
+  --evaluationId 42 \
+  --apiKey $EVALAI_API_KEY
+evalai check
+Option	Description
+--evaluationId <id>	Required. Evaluation to gate on
+--apiKey <key>	API key (or EVALAI_API_KEY)
+--format <fmt>	human, json, or github
+--onFail import	Import failing run to dashboard
+--explain	Show score breakdown
+--minScore <n>	Fail if score < n
+--warnDrop <n>	Warn if regression exceeds n
+--maxDrop <n>	Fail if regression exceeds n
+--minN <n>	Fail if test count < n
+--allowWeakEvidence	Permit weak evidence
+--policy <name>	HIPAA, SOC2, GDPR, PCI_DSS, FINRA_4511
+--baseline <mode>	published, previous, production
+--fail-on-flake	Fail if any case is flaky
+--baseUrl <url>	Override API base URL
+
+Exit codes
+Code	Meaning
+0	PASS
+8	WARN
+1	Score below threshold
+2	Regression failure
+3	Policy violation
+4	API error
+5	Bad arguments
+6	Low test count
+7	Weak evidence
+evalai doctor
+Verify CI setup before running the gate:
+
+npx -y @pauly4010/evalai-sdk@^1 doctor \
+  --evaluationId 42 \
+  --apiKey $EVALAI_API_KEY
+If doctor passes, check will work.
+
+🧯 Error Handling
+import { EvalAIError, RateLimitError } from "@pauly4010/evalai-sdk";
 
 try {
-  await client.traces.create({...});
-} catch (error) {
-  if (error instanceof RateLimitError) {
-    console.log('Rate limited, retry after:', error.retryAfter);
-  } else if (error instanceof EvalAIError) {
-    console.log('Error:', error.code, error.message);
+  await client.traces.create({ name: "User Query" });
+} catch (err) {
+  if (err instanceof RateLimitError) {
+    console.log("Retry after:", err.retryAfter);
+  } else if (err instanceof EvalAIError) {
+    console.log(err.code, err.message, err.requestId);
   }
 }
-```
 
-## Advanced Features
 
-### Context Propagation
-
-```typescript
-import { withContext } from "@pauly4010/evalai-sdk";
-
-withContext({ userId: "123", sessionId: "abc" }, async () => {
-  // Context automatically included in all traces
-  await client.traces.create({
-    name: "Query",
-    traceId: "trace-1",
-  });
+🔍 Traces
+const trace = await client.traces.create({
+  name: "User Query",
+  traceId: "trace-123",
+  metadata: { userId: "456" },
 });
-```
 
-### Test Suites
 
-```typescript
-import { createTestSuite } from "@pauly4010/evalai-sdk";
+📝 Evaluations
+import { EvaluationTemplates } from "@pauly4010/evalai-sdk";
 
-const suite = createTestSuite({
-  name: "Chatbot Tests",
-  tests: [
-    {
-      name: "Greeting",
-      input: "Hello",
-      expectedOutput: "Hi there!",
-    },
-  ],
+const evaluation = await client.evaluations.create({
+  name: "Chatbot Responses",
+  type: EvaluationTemplates.OUTPUT_QUALITY,
+  createdBy: userId,
 });
 
-await suite.run(client);
-```
 
-### Framework Integrations
-
-```typescript
+🔌 Framework Integrations
 import { traceOpenAI } from "@pauly4010/evalai-sdk/integrations/openai";
 import OpenAI from "openai";
 
 const openai = traceOpenAI(new OpenAI(), client);
 
-// All OpenAI calls are automatically traced
-const response = await openai.chat.completions.create({
+await openai.chat.completions.create({
   model: "gpt-4",
   messages: [{ role: "user", content: "Hello" }],
 });
-```
-
-## TypeScript Support
-
-The SDK is fully typed with TypeScript generics for type-safe metadata:
-
-```typescript
-interface CustomMetadata {
-  userId: string;
-  sessionId: string;
-  model: string;
-}
-
-const trace = await client.traces.create<CustomMetadata>({
-  name: "Query",
-  traceId: "trace-1",
-  metadata: {
-    userId: "123",
-    sessionId: "abc",
-    model: "gpt-4",
-  },
-});
-
-// TypeScript knows the exact metadata type
-console.log(trace.metadata.userId);
-```
-
-## 📋 Annotations API (v1.2.0)
-
-Human-in-the-loop evaluation for quality assurance:
-
-```typescript
-// Create an annotation
-const annotation = await client.annotations.create({
-  evaluationRunId: 123,
-  testCaseId: 456,
-  rating: 5,
-  feedback: "Excellent response!",
-  labels: { category: "helpful", sentiment: "positive" },
-});
-
-// List annotations
-const annotations = await client.annotations.list({
-  evaluationRunId: 123,
-});
-
-// Annotation Tasks
-const task = await client.annotations.tasks.create({
-  name: "Q4 Quality Review",
-  type: "classification",
-  organizationId: 1,
-  instructions: "Rate responses from 1-5",
-});
-
-const tasks = await client.annotations.tasks.list({
-  organizationId: 1,
-  status: "pending",
-});
-
-const taskDetail = await client.annotations.tasks.get(taskId);
-
-// Annotation Items
-const item = await client.annotations.tasks.items.create(taskId, {
-  content: "Response to evaluate",
-  annotation: { rating: 4, category: "good" },
-});
-
-const items = await client.annotations.tasks.items.list(taskId);
-```
-
-## 🔑 Developer API (v1.2.0)
-
-Manage API keys, webhooks, and monitor usage:
-
-### API Keys
-
-```typescript
-// Create an API key
-const { apiKey, id, keyPrefix } = await client.developer.apiKeys.create({
-  name: "Production Key",
-  organizationId: 1,
-  scopes: ["traces:read", "traces:write", "evaluations:read"],
-  expiresAt: "2025-12-31T23:59:59Z",
-});
 
-// IMPORTANT: Save the apiKey securely - it's only shown once!
 
-// List API keys
-const keys = await client.developer.apiKeys.list({
-  organizationId: 1,
-});
+🧭 Changelog
+v1.5.5 (Latest)
+PASS/WARN/FAIL gate semantics
 
-// Update an API key
-await client.developer.apiKeys.update(keyId, {
-  name: "Updated Name",
-  scopes: ["traces:read"],
-});
+--warnDrop soft regression band
 
-// Revoke an API key
-await client.developer.apiKeys.revoke(keyId);
+Flake intelligence + per-case pass rates
 
-// Get usage statistics for a key
-const usage = await client.developer.apiKeys.getUsage(keyId);
-console.log("Total requests:", usage.totalRequests);
-console.log("By endpoint:", usage.usageByEndpoint);
-```
+--fail-on-flake enforcement
 
-### Webhooks
+Golden regression suite
 
-```typescript
-// Create a webhook
-const webhook = await client.developer.webhooks.create({
-  organizationId: 1,
-  url: "https://your-app.com/webhooks/evalai",
-  events: ["trace.created", "evaluation.completed", "annotation.created"],
-});
+Nightly determinism + performance audits
 
-// List webhooks
-const webhooks = await client.developer.webhooks.list({
-  organizationId: 1,
-  status: "active",
-});
+Audit trail, observability, retention, and migration safety docs
 
-// Get a specific webhook
-const webhookDetail = await client.developer.webhooks.get(webhookId);
+v1.5.0
+GitHub annotations formatter
 
-// Update a webhook
-await client.developer.webhooks.update(webhookId, {
-  url: "https://new-url.com/webhooks",
-  events: ["trace.created"],
-  status: "inactive",
-});
+JSON formatter
 
-// Delete a webhook
-await client.developer.webhooks.delete(webhookId);
+--onFail import
 
-// Get webhook deliveries (for debugging)
-const deliveries = await client.developer.webhooks.getDeliveries(webhookId, {
-  limit: 50,
-  success: false, // Only failed deliveries
-});
-```
+--explain
 
-### Usage Analytics
+evalai doctor
 
-```typescript
-// Get detailed usage statistics
-const stats = await client.developer.getUsage({
-  organizationId: 1,
-  startDate: "2025-01-01",
-  endDate: "2025-01-31",
-});
+CI pinned invocation guidance
 
-console.log("Traces:", stats.traces.total);
-console.log("Evaluations by type:", stats.evaluations.byType);
-console.log("API calls by endpoint:", stats.apiCalls.byEndpoint);
 
-// Get usage summary
-const summary = await client.developer.getUsageSummary(organizationId);
-console.log("Current period:", summary.currentPeriod);
-console.log("Limits:", summary.limits);
-```
-
-## ⚖️ LLM Judge Extended (v1.2.0)
+Environment Variable Safety
 
-Enhanced LLM judge configuration and analysis:
-
-```typescript
-// Create a judge configuration
-const config = await client.llmJudge.createConfig({
-  name: "GPT-4 Accuracy Judge",
-  description: "Evaluates factual accuracy",
-  model: "gpt-4",
-  rubric: "Score 1-10 based on factual accuracy...",
-  temperature: 0.3,
-  maxTokens: 500,
-  organizationId: 1,
-  createdBy: userId,
-});
+The SDK never assumes `process.env` exists. All environment reads are guarded, so the client can initialize safely in browser, edge, and server runtimes.
 
-// List configurations
-const configs = await client.llmJudge.listConfigs({
-  organizationId: 1,
-});
+If environment variables are unavailable, the SDK falls back to explicit config.
 
-// List results
-const results = await client.llmJudge.listResults({
-  configId: config.id,
-  evaluationId: 123,
-});
-
-// Get alignment analysis
-const alignment = await client.llmJudge.getAlignment({
-  configId: config.id,
-  startDate: "2025-01-01",
-  endDate: "2025-01-31",
-});
-
-console.log("Average score:", alignment.averageScore);
-console.log("Accuracy:", alignment.alignmentMetrics.accuracy);
-console.log("Agreement with human:", alignment.comparisonWithHuman?.agreement);
-```
-
-## 🏢 Organizations API (v1.2.0)
-
-Manage organization details:
-
-```typescript
-// Get current organization
-const org = await client.organizations.getCurrent();
-console.log("Organization:", org.name);
-console.log("Plan:", org.plan);
-console.log("Status:", org.status);
-```
-
-## evalai CLI (v1.4.1)
-
-The SDK includes a CLI for CI/CD evaluation gates. Install globally or use via `npx`:
-
-```bash
-# Via npx (no global install)
-npx @pauly4010/evalai-sdk check --minScore 92 --evaluationId 42 --apiKey $EVALAI_API_KEY
-
-# Or install globally
-npm install -g @pauly4010/evalai-sdk
-evalai check --minScore 92 --evaluationId 42
-```
-
-### evalai check
-
-Gate deployments on quality scores, regression, and compliance:
-
-| Option | Description |
-|--------|-------------|
-| `--evaluationId <id>` | **Required.** Evaluation to gate on |
-| `--apiKey <key>` | API key (or `EVALAI_API_KEY` env) |
-| `--minScore <n>` | Fail if score &lt; n (0–100) |
-| `--maxDrop <n>` | Fail if score dropped &gt; n from baseline |
-| `--minN <n>` | Fail if total test cases &lt; n |
-| `--allowWeakEvidence` | Permit weak evidence level |
-| `--policy <name>` | Enforce HIPAA, SOC2, GDPR, PCI_DSS, FINRA_4511 |
-| `--baseline <mode>` | `published`, `previous`, or `production` |
-| `--baseUrl <url>` | API base URL |
-
-**Exit codes:** 0=pass, 1=score below, 2=regression, 3=policy violation, 4=API error, 5=bad args, 6=low N, 7=weak evidence
-
-## Changelog
-
-### v1.4.1 (Latest)
-
-- **evalai check `--baseline production`** — Compare against latest prod-tagged run
-- **Package hardening** — Leaner npm publish with `files`, `sideEffects: false`
-
-### v1.4.0
-
-- **evalai CLI** — Command-line tool for CI/CD evaluation gates
-  - `evalai check` — Gate deployments on quality scores, regression, and compliance
-  - `--minScore <n>` — Fail if quality score &lt; n (0–100)
-  - `--maxDrop <n>` — Fail if score dropped &gt; n points from baseline
-  - `--minN <n>` — Fail if total test cases &lt; n
-  - `--allowWeakEvidence` — Permit weak evidence level (default: fail)
-  - `--policy <name>` — Enforce compliance (HIPAA, SOC2, GDPR, PCI_DSS, FINRA_4511)
-  - `--baseline <mode>` — Compare to `published` or `previous` run
-  - `--evaluationId <id>` — Required. Evaluation to gate on
-  - Environment: `EVALAI_API_KEY`, `EVALAI_BASE_URL`
-  - Exit codes: 0=pass, 1=score below, 2=regression, 3=policy violation, 4=API error, 5=bad args, 6=low N, 7=weak evidence
-- **CLI Exports** — `parseArgs`, `runCheck`, `EXIT` from `@pauly4010/evalai-sdk` for programmatic use
-
-### v1.3.0
-
-- **Workflow Tracing** — Multi-agent orchestration with full lifecycle instrumentation
-  - `WorkflowTracer` class with `startWorkflow`, `endWorkflow`, `startAgentSpan`, `endAgentSpan`
-  - `createWorkflowTracer` convenience factory
-  - `traceWorkflowStep` generic wrapper for any async function
-  - Agent handoff recording (`delegation`, `escalation`, `parallel`, `fallback`)
-  - Decision auditing with alternatives, confidence scores, reasoning, and context factors
-  - Cost tracking per span/workflow with automatic pricing (16+ models)
-  - Cost breakdown by category (`llm`, `tool`, `embedding`, `other`)
-- **Framework Integrations** — Wrap popular multi-agent frameworks:
-  - `traceLangChainAgent` — wraps `.invoke()` and `.call()` with auto-tracing
-  - `traceCrewAI` — wraps `.kickoff()` with workflow start/end
-  - `traceAutoGen` — wraps `.initiate_chat()` with workflow start/end
-- **Performance Utilities**
-  - `RequestCache` with configurable TTL (`CacheTTL` presets)
-  - `PaginatedIterator` / `createPaginatedIterator` / `autoPaginate` for cursor-based pagination
-  - `RequestBatcher` for batching API calls
-  - `RateLimiter` client-side rate limit handling
-- **Cost Tracking Types** — `CostRecord`, `CostBreakdown`, `ProviderPricing` interfaces
-- **Agent Decision Auditing Types** — `AgentDecision`, `DecisionAlternative`, `RecordDecisionParams` interfaces
-- **Benchmark Types** — `Benchmark`, `BenchmarkResult`, `AgentConfig` interfaces
-
-### v1.2.1 (Bug Fixes)
-
-- 🐛 **Critical Fixes**
-  - Fixed CLI import paths for proper npm package distribution
-  - Fixed duplicate trace creation in OpenAI/Anthropic integrations
-  - Fixed Commander.js command structure
-  - Added browser/Node.js environment detection and helpful errors
-  - Fixed context system to work in both Node.js and browsers
-  - Added security checks to snapshot path sanitization
-  - Removed misleading empty exports (StreamingClient, BatchClient)
-- 📦 **Dependencies**
-  - Updated Commander to v14
-  - Added peer dependencies for OpenAI and Anthropic SDKs (optional)
-  - Added Node.js engine requirement (>=16.0.0)
-- 📚 **Documentation**
-  - Clarified Node.js-only vs universal features
-  - Added environment support section
-  - Updated examples with security best practices
-
-### v1.2.0
-
-- 🎉 **100% API Coverage** - All backend endpoints now supported!
-- 📋 **Annotations API** - Complete human-in-the-loop evaluation
-  - Create and list annotations
-  - Manage annotation tasks
-  - Handle annotation items
-- 🔑 **Developer API** - Full API key and webhook management
-  - CRUD operations for API keys
-  - Webhook management with delivery tracking
-  - Usage analytics and monitoring
-- ⚖️ **LLM Judge Extended** - Enhanced judge capabilities
-  - Configuration management
-  - Results querying
-  - Alignment analysis
-- 🏢 **Organizations API** - Organization details access
-- 📊 **Enhanced Types** - 40+ new TypeScript interfaces
-- 📚 **Comprehensive Documentation** - Examples for all new features
-
-### v1.1.0
-
-- ✨ Added comprehensive evaluation template types
-- ✨ Added organization resource limits tracking
-- ✨ Added `getOrganizationLimits()` method
-- 📚 Enhanced documentation with new features
-
-### v1.0.0
-
-- 🎉 Initial release
-- ✅ Traces, Evaluations, LLM Judge APIs
-- ✅ Framework integrations (OpenAI, Anthropic)
-- ✅ Test suite builder
-- ✅ Context propagation
-- ✅ Error handling & retries
-
-## License
 
+📄 License
 MIT
 
-## Support
+🤝 Support
+Documentation:
+https://v0-ai-evaluation-platform-nu.vercel.app/documentation
 
-- Documentation: https://v0-ai-evaluation-platform-nu.vercel.app/documentation
-- Issues: https://github.com/pauly7610/ai-evaluation-platform/issues
+Issues:
+https://github.com/pauly7610/ai-evaluation-platform/issues
+```