How to Connect Pinecone to an AI Agent
Auth setup
1. Go to https://app.pinecone.io and create a project. 2. Copy the API key from the API Keys tab. 3. For data ops you also need the index host URL — get it from the console or from GET /indexes/{name}.
Key facts
| Base URL | https://api.pinecone.io/ |
|---|
| API version | 2024-10 (control plane) + index-specific data-plane URLs |
|---|
| Auth | API key authentication via the Api-Key header (NOT Authorization). Each Pinecone project has its own API keys. The control plane (api.pinecone.io) manages indexes; the data plane has per-index URLs like https://{index-name}-{project-id}.svc.{environment}.pinecone.io/. |
|---|
| Scopes | Project-scoped (one API key controls all indexes in the project). |
|---|
| Request body | application/json |
|---|
| Pagination | Query results return topK matches; for listing operations, response may include a `pagination.next` token. |
|---|
| Rate limit | Serverless: scales automatically, billed per operation (no hard per-minute cap). Pod-based: limited by pod size and replica count. Control plane: ~100 requests/min for admin endpoints. |
|---|
| Error format | JSON: {"code":3,"message":"...","details":[...]} — gRPC-style codes for the data plane. |
|---|
Key endpoints
| Method | Path | Description |
|---|
GET | /indexes | List indexes in the project |
POST | /indexes | Create a new serverless or pod-based index |
GET | /indexes/{index_name} | Get index metadata including data-plane host |
POST | /vectors/upsert | [data-plane] Upsert vectors to an index |
POST | /query | [data-plane] Similarity search |
POST | /vectors/delete | [data-plane] Delete vectors by ID or filter |
Quickstart
POST https://{index-host}/query
Api-Key: {api_key}
Content-Type: application/json
{"vector":[0.1,0.2,...],"topK":5,"includeMetadata":true,"namespace":"default"}
Response: {"matches":[{"id":"v1","score":0.92,"metadata":{...}}],"namespace":"default"}
Agent pitfalls & tips
- Data plane and control plane use DIFFERENT base URLs. Always call GET /indexes/{name} first to resolve the data-plane host, then cache it.
- The auth header is `Api-Key: {key}` — it is NOT Authorization/Bearer. This trips up most generic HTTP clients.
- Vector dimension must match the index dimension exactly — a mismatch returns 400 with "Vector dimension X does not match index dimension Y".
- Use namespaces to partition data within a single index — much cheaper than creating multiple indexes.
- Serverless indexes are cheaper for sparse / bursty workloads; pod-based is cheaper for sustained high QPS.
- The Pinecone MCP server (`@pinecone-database/mcp`) exposes query/upsert/delete with the key+host pre-configured.
Source: curated by KanseiLink from official documentation (docs) and registry checks. Last reviewed: 2026-04-10. Specs change — verify against the official docs before production use.
Frequently Asked Questions
What is Pinecone's AEO score?
▼
Pinecone has an AEO score of 0.70 and is rated A (Functional agent integration). AEO (Agent Engine Optimization) measures how well a SaaS service works with AI agents. Scores range from 0.00 to 1.00, with grades from AAA (best) to D (not agent-ready).
Is Pinecone AI-agent-ready?
▼
Pinecone is currently connectable for AI agent use. Third-party MCP integrations are available for this service. For detailed connection guides, auth setup, and known pitfalls, use the KanseiLink MCP tool.
How does Pinecone compare to other AI & ML services?
▼
In the AI & ML category, Pinecone is rated A. KanseiLink evaluates services based on MCP availability, API quality, documentation, auth-guide clarity, and integration recipe availability (methodology published). Visit the full rankings at kansei-link.com to see how Pinecone compares.
How can I integrate Pinecone with an AI agent?
▼
The fastest way to integrate Pinecone with an AI agent is through KanseiLink MCP. Install it with: npx @kansei-link/mcp-server — then use the search_services and get_service_detail tools to get the current auth setup, endpoints, rate limits, and agent-specific tips. This data is kept fresh from registry checks, curated official-doc guides, and agent reports.
How do I authenticate with Pinecone?
▼
API key authentication via the Api-Key header (NOT Authorization). Each Pinecone project has its own API keys. The control plane (api.pinecone.io) manages indexes; the data plane has per-index URLs like https://{index-name}-{project-id}.svc.{environment}.pinecone.io/. Setup: 1. Go to https://app.pinecone.io and create a project. 2. Copy the API key from the API Keys tab. 3. For data ops you also need the index host URL — get it from the console or from GET /indexes/{name}.
What are Pinecone's API rate limits?
▼
Serverless: scales automatically, billed per operation (no hard per-minute cap). Pod-based: limited by pod size and replica count. Control plane: ~100 requests/min for admin endpoints.