feat(mcp-server): Add 7 missing extraction async jobs tools

- extraction.extractBatch: batch extraction with shared config
- extraction.submitJob: async job submission with webhook support
- extraction.getJob: get job status/results by ID
- extraction.listJobs: list recent async jobs
- extraction.getProductRateLimitStatus: per-product or summary rate limits
- extraction.resetProductRateLimit: admin rate limit reset
- extraction.sidecarMonitoringState: detailed sidecar circuit breaker state

All tools require admin role and map to existing extraction-service endpoints.
Fixes TypeScript optional parameter error in extractionGetProductRateLimitStatus.

services/mcp-server/src/lib/extraction-client.ts

@@ -70,3 +70,163 @@ export async function extractionSidecarHealth(opts: { requestId?: string }): Pro
   const res = await fetch(url, { headers, signal: AbortSignal.timeout(10_000) });
   return res.json();
 }
+
+export async function extractionExtractBatch(
+  params: {
+    inputs: Array<{
+      text: string;
+      taskId?: string;
+      taskPrompt?: string;
+    }>;
+    examples?: Array<{
+      text: string;
+      extractions: ExtractionItem[];
+    }>;
+    modelId?: string;
+  },
+  opts: { requestId?: string }
+): Promise<unknown> {
+  const url = `${config.EXTRACTION_SERVICE_URL}/api/extract/batch`;
+  const headers: Record<string, string> = {
+    'Content-Type': 'application/json',
+    ...(opts.requestId ? { 'x-request-id': opts.requestId } : {}),
+  };
+  const res = await fetch(url, {
+    method: 'POST',
+    headers,
+    body: JSON.stringify(params),
+    signal: AbortSignal.timeout(30_000),
+  });
+  if (!res.ok) {
+    const body = await res.text().catch(() => '');
+    throw new Error(`extraction-service POST /api/extract/batch → ${res.status}: ${body}`);
+  }
+  return res.json();
+}
+
+export async function extractionSubmitJob(
+  params: {
+    inputs: Array<{
+      text: string;
+      taskId?: string;
+      taskPrompt?: string;
+    }>;
+    examples?: Array<{
+      text: string;
+      extractions: ExtractionItem[];
+    }>;
+    modelId?: string;
+    productId?: string;
+    webhookUrl?: string;
+    webhookSecret?: string;
+    webhookRetryAttempts?: number;
+  },
+  opts: { requestId?: string }
+): Promise<unknown> {
+  const url = `${config.EXTRACTION_SERVICE_URL}/api/extract/jobs`;
+  const headers: Record<string, string> = {
+    'Content-Type': 'application/json',
+    ...(opts.requestId ? { 'x-request-id': opts.requestId } : {}),
+  };
+  const res = await fetch(url, {
+    method: 'POST',
+    headers,
+    body: JSON.stringify(params),
+    signal: AbortSignal.timeout(30_000),
+  });
+  if (!res.ok) {
+    const body = await res.text().catch(() => '');
+    throw new Error(`extraction-service POST /api/extract/jobs → ${res.status}: ${body}`);
+  }
+  return res.json();
+}
+
+export async function extractionGetJob(
+  jobId: string,
+  opts: { requestId?: string }
+): Promise<unknown> {
+  const url = `${config.EXTRACTION_SERVICE_URL}/api/extract/jobs/${jobId}`;
+  const headers: Record<string, string> = {
+    ...(opts.requestId ? { 'x-request-id': opts.requestId } : {}),
+  };
+  const res = await fetch(url, { headers, signal: AbortSignal.timeout(10_000) });
+  if (!res.ok) {
+    const body = await res.text().catch(() => '');
+    throw new Error(`extraction-service GET /api/extract/jobs/${jobId} → ${res.status}: ${body}`);
+  }
+  return res.json();
+}
+
+export async function extractionListJobs(opts: { requestId?: string }): Promise<unknown> {
+  const url = `${config.EXTRACTION_SERVICE_URL}/api/extract/jobs`;
+  const headers: Record<string, string> = {
+    ...(opts.requestId ? { 'x-request-id': opts.requestId } : {}),
+  };
+  const res = await fetch(url, { headers, signal: AbortSignal.timeout(10_000) });
+  if (!res.ok) {
+    const body = await res.text().catch(() => '');
+    throw new Error(`extraction-service GET /api/extract/jobs → ${res.status}: ${body}`);
+  }
+  return res.json();
+}
+
+export async function extractionGetProductRateLimitStatus(
+  productId?: string,
+  opts?: { requestId?: string }
+): Promise<unknown> {
+  const url = productId
+    ? `${config.EXTRACTION_SERVICE_URL}/api/extract/rate-limits/product?productId=${encodeURIComponent(productId)}`
+    : `${config.EXTRACTION_SERVICE_URL}/api/extract/rate-limits/product`;
+  const headers: Record<string, string> = {
+    ...(opts?.requestId ? { 'x-request-id': opts.requestId } : {}),
+  };
+  const res = await fetch(url, { headers, signal: AbortSignal.timeout(10_000) });
+  if (!res.ok) {
+    const body = await res.text().catch(() => '');
+    throw new Error(
+      `extraction-service GET /api/extract/rate-limits/product → ${res.status}: ${body}`
+    );
+  }
+  return res.json();
+}
+
+export async function extractionResetProductRateLimit(
+  productId: string,
+  opts: { requestId?: string }
+): Promise<unknown> {
+  const url = `${config.EXTRACTION_SERVICE_URL}/api/extract/rate-limits/product/reset`;
+  const headers: Record<string, string> = {
+    'Content-Type': 'application/json',
+    ...(opts.requestId ? { 'x-request-id': opts.requestId } : {}),
+  };
+  const res = await fetch(url, {
+    method: 'POST',
+    headers,
+    body: JSON.stringify({ productId }),
+    signal: AbortSignal.timeout(10_000),
+  });
+  if (!res.ok) {
+    const body = await res.text().catch(() => '');
+    throw new Error(
+      `extraction-service POST /api/extract/rate-limits/product/reset → ${res.status}: ${body}`
+    );
+  }
+  return res.json();
+}
+
+export async function extractionSidecarMonitoringState(opts: {
+  requestId?: string;
+}): Promise<unknown> {
+  const url = `${config.EXTRACTION_SERVICE_URL}/api/extract/monitoring/sidecar`;
+  const headers: Record<string, string> = {
+    ...(opts.requestId ? { 'x-request-id': opts.requestId } : {}),
+  };
+  const res = await fetch(url, { headers, signal: AbortSignal.timeout(10_000) });
+  if (!res.ok) {
+    const body = await res.text().catch(() => '');
+    throw new Error(
+      `extraction-service GET /api/extract/monitoring/sidecar → ${res.status}: ${body}`
+    );
+  }
+  return res.json();
+}

services/mcp-server/src/modules/extraction/extraction-tools.ts

@@ -5,6 +5,13 @@ import {
   extractionModels,
   extractionCacheStats,
   extractionSidecarHealth,
+  extractionExtractBatch,
+  extractionSubmitJob,
+  extractionGetJob,
+  extractionListJobs,
+  extractionGetProductRateLimitStatus,
+  extractionResetProductRateLimit,
+  extractionSidecarMonitoringState,
 } from '../../lib/extraction-client.js';
 
 registerTool({
@@ -55,3 +62,147 @@ registerTool({
     return extractionSidecarHealth({ requestId: req.id });
   },
 });
+
+registerTool({
+  name: 'extraction.extractBatch',
+  description:
+    'Run batch extraction on multiple inputs with shared configuration. Returns array of extraction results. Requires admin role.',
+  requiredRole: 'admin',
+  inputSchema: z.object({
+    inputs: z
+      .array(
+        z.object({
+          text: z.string().min(1).describe('Text to extract from'),
+          taskId: z.string().optional().describe('Extraction task ID'),
+          taskPrompt: z.string().optional().describe('Custom task prompt'),
+        })
+      )
+      .min(1)
+      .describe('Array of extraction inputs'),
+    examples: z
+      .array(
+        z.object({
+          text: z.string().min(1).describe('Example text'),
+          extractions: z.array(
+            z.object({
+              extraction_class: z.string(),
+              extraction_text: z.string(),
+              attributes: z.record(z.string()).optional(),
+            })
+          ),
+        })
+      )
+      .optional()
+      .describe('Few-shot examples'),
+    modelId: z.string().optional().describe('Override model ID'),
+  }),
+  async execute(args, req) {
+    return extractionExtractBatch(args, { requestId: req.id });
+  },
+});
+
+registerTool({
+  name: 'extraction.submitJob',
+  description:
+    'Submit an async batch extraction job. Returns jobId for polling. Supports webhook callbacks. Requires admin role.',
+  requiredRole: 'admin',
+  inputSchema: z.object({
+    inputs: z
+      .array(
+        z.object({
+          text: z.string().min(1).describe('Text to extract from'),
+          taskId: z.string().optional().describe('Extraction task ID'),
+          taskPrompt: z.string().optional().describe('Custom task prompt'),
+        })
+      )
+      .min(1)
+      .describe('Array of extraction inputs'),
+    examples: z
+      .array(
+        z.object({
+          text: z.string().min(1).describe('Example text'),
+          extractions: z.array(
+            z.object({
+              extraction_class: z.string(),
+              extraction_text: z.string(),
+              attributes: z.record(z.string()).optional(),
+            })
+          ),
+        })
+      )
+      .optional()
+      .describe('Few-shot examples'),
+    modelId: z.string().optional().describe('Override model ID'),
+    productId: z.string().optional().describe('Product ID for rate limiting'),
+    webhookUrl: z.string().url().optional().describe('Webhook URL for job completion'),
+    webhookSecret: z.string().optional().describe('Webhook secret for HMAC validation'),
+    webhookRetryAttempts: z
+      .number()
+      .int()
+      .min(0)
+      .max(10)
+      .optional()
+      .describe('Webhook retry attempts'),
+  }),
+  async execute(args, req) {
+    return extractionSubmitJob(args, { requestId: req.id });
+  },
+});
+
+registerTool({
+  name: 'extraction.getJob',
+  description: 'Get status and results of an async extraction job by ID. Requires admin role.',
+  requiredRole: 'admin',
+  inputSchema: z.object({
+    jobId: z.string().min(1).describe('Job ID to retrieve'),
+  }),
+  async execute(args, req) {
+    return extractionGetJob(args.jobId, { requestId: req.id });
+  },
+});
+
+registerTool({
+  name: 'extraction.listJobs',
+  description: 'List recent async extraction jobs. Requires admin role.',
+  requiredRole: 'admin',
+  inputSchema: z.object({}),
+  async execute(_args, req) {
+    return extractionListJobs({ requestId: req.id });
+  },
+});
+
+registerTool({
+  name: 'extraction.getProductRateLimitStatus',
+  description:
+    'Get product rate limit status. Pass productId for specific product, omit for summary of all products. Requires admin role.',
+  requiredRole: 'admin',
+  inputSchema: z.object({
+    productId: z.string().optional().describe('Product ID to check (omit for all products)'),
+  }),
+  async execute(args, req) {
+    return extractionGetProductRateLimitStatus(args.productId, { requestId: req.id });
+  },
+});
+
+registerTool({
+  name: 'extraction.resetProductRateLimit',
+  description: 'Reset rate limit for a specific product (admin operation). Requires admin role.',
+  requiredRole: 'admin',
+  inputSchema: z.object({
+    productId: z.string().min(1).describe('Product ID to reset rate limit for'),
+  }),
+  async execute(args, req) {
+    return extractionResetProductRateLimit(args.productId, { requestId: req.id });
+  },
+});
+
+registerTool({
+  name: 'extraction.sidecarMonitoringState',
+  description:
+    'Get detailed sidecar health monitoring state and circuit breaker information. Requires admin role.',
+  requiredRole: 'admin',
+  inputSchema: z.object({}),
+  async execute(_args, req) {
+    return extractionSidecarMonitoringState({ requestId: req.id });
+  },
+});