{
  "markdown": "# Pinecone Plugin for Claude Code\n\nA lightweight plugin that integrates [Pinecone](https://www.pinecone.io/) vector database capabilities directly into Claude Code, enabling semantic search, index management, and RAG (Retrieval Augmented Generation) workflows.\n\n## Features\n\n- **Pinecone Assistant** – Fully managed RAG service for document Q&A with citations, natural language support, and incremental file syncing\n- **Pinecone MCP Server** – Full integration with the Pinecone Model Context Protocol server for index creation, listing, searching, and more\n- **Slash Commands** – Quick access to common Pinecone operations directly from Claude Code\n- **Semantic Search** – Query your vector indexes using natural language\n- **Natural Language Recognition** – Assistant commands work without explicit slash commands\n\n## Installation\n\n### Option A: Claude Code Plugins Directory (Recommended)\n\nInstall from the [official Claude Code Plugins Directory](https://github.com/anthropics/claude-plugins-official):\n\n1. Install the plugin:\n   ```\n   /plugin install pinecone\n   ```\n\n2. **Restart Claude Code** to activate the plugin.\n\n### Option B: Pinecone Marketplace\n\nAlternatively, install directly from the Pinecone marketplace:\n\n1. **Add the Pinecone plugin marketplace:**\n   ```\n   /plugin marketplace add pinecone-io/pinecone-claude-code-plugin\n   ```\n\n2. **Install the plugin:**\n   ```\n   /plugin install pinecone@pinecone-claude-code-plugin\n   ```\n\n3. **When prompted**, select your preferred installation scope:\n   - **User scope** (default) – Available across all your projects\n   - **Project scope** – Shared with your team via version control\n   - **Local scope** – Project-specific, not shared (gitignored)\n\n4. **Restart Claude Code** to activate the plugin.\n\n### Set Your API Key\n\nAfter installing via either method, configure your Pinecone API key before running Claude Code:\n\n```bash\nexport PINECONE_API_KEY=\"your-api-key-here\"\n```\n\n> **Don't have a Pinecone account?** Sign up for free at [app.pinecone.io](https://app.pinecone.io/?sessionType=signup)\n\n\n### Install uv (Required for Assistant Commands)\n\nTo use Pinecone Assistant functionality, you must have uv installed. uv is a fast Python package and project manager:\n\n**macOS and Linux:**\n```bash\ncurl -LsSf https://astral.sh/uv/install.sh | sh\n```\n\n**Windows:**\n```powershell\npowershell -ExecutionPolicy ByPass -c \"irm https://astral.sh/uv/install.ps1 | iex\"\n```\n\n**With Homebrew:**\n```bash\nbrew install uv\n```\n\nAfter installation, restart your terminal and verify with: `uv --version`\n\nFull installation guide: https://docs.astral.sh/uv/getting-started/installation/\n\n### Install the Pinecone CLI (Optional)\n\nFor additional command-line capabilities, install the Pinecone CLI:\n\n```bash\nbrew tap pinecone-io/tap\nbrew install pinecone-io/tap/pinecone\n```\n\n## Available Skills\n\n### `/pinecone:help`\n\nOverview of all available Pinecone skills and what you need to get started. Run this when first installing the plugin.\n\n### `/pinecone:quickstart`\n\nInteractive quickstart for new developers. Choose between two paths:\n- **Database** — Create an integrated index, upsert data, and query using Pinecone MCP + Python\n- **Assistant** — Create a Pinecone Assistant for document Q&A with citations\n\n### `/pinecone:query`\n\nQuery integrated indexes using natural language. Wraps the Pinecone MCP server for easy searching.\n\n```\n/pinecone:query query [your search text] index [indexName] namespace [ns] reranker [rerankModel]\n```\n\n> **Note:** Only works with integrated indexes that use Pinecone's hosted embedding models.\n\n### `/pinecone:full-text-search`\n\nEnd-to-end workflow for Pinecone's full-text-search (FTS) preview API (`2026-01.alpha`) — design a document schema, ingest a corpus, and construct `documents.search(...)` calls. Covers BM25 (`text` / `query_string`), `dense_vector` and `sparse_vector` scoring, and text-match filters (`$match_phrase` / `$match_all` / `$match_any`) for hybrid lexical+semantic queries.\n\nShips a `scripts/ingest.py` helper that does bulk `batch_upsert` with per-batch error inspection and post-upsert readiness polling — the three things bare-LLM ingest code reliably skips.\n\n> **Requires `pinecone` Python SDK ≥ 9.0.** The FTS document-schema API lives under `pinecone.preview`.\n\n### `/pinecone:assistant`\n\nAll-in-one skill for Pinecone Assistants — create, upload, sync, chat, context retrieval, and list. Works with both slash commands and natural language:\n\n- \"Create a Pinecone assistant from my docs\"\n- \"Upload files from ./docs to my-assistant\"\n- \"Sync my assistant with the docs folder\"\n- \"Ask my assistant about authentication\"\n- \"Search my assistant for context about embeddings\"\n\n**Learn more:** https://docs.pinecone.io/guides/assistant/quickstart\n\n### `/pinecone:cli`\n\nGuide for using the Pinecone CLI (`pc`) to manage resources from the terminal. The CLI supports all index types and vector operations.\n\n### `/pinecone:mcp`\n\nReference for all Pinecone MCP server tools — parameters, usage, and examples.\n\n### `/pinecone:docs`\n\nCurated documentation reference with links to official docs organized by topic and data format references.\n\n## MCP Server Tools\n\nThe plugin includes the full Pinecone MCP Server with the following tools:\n\n| Tool | Description |\n|------|-------------|\n| `list-indexes` | List all available Pinecone indexes |\n| `describe-index` | Get index configuration and namespaces |\n| `describe-index-stats` | Get statistics including record counts and namespaces |\n| `search-records` | Search records with optional metadata filtering and reranking |\n| `create-index-for-model` | Create a new index with integrated embeddings |\n| `upsert-records` | Insert or update records in an index |\n| `rerank-documents` | Rerank documents using a specified reranking model |\n\nFor complete MCP server documentation, visit: [Pinecone MCP Server Guide](https://docs.pinecone.io/guides/operations/mcp-server)\n\n## Troubleshooting\n\n### \"API Key not found\" or access errors\n\nMake sure your `PINECONE_API_KEY` environment variable is set correctly:\n\n```bash\necho $PINECONE_API_KEY\n```\n\nIf it's empty, set it and restart Claude Code.\n\n### MCP server not responding\n\n1. Ensure you have Node.js installed (the MCP server runs via `npx`)\n2. Check that your API key is valid\n3. Restart Claude Code after setting environment variables\n\n### Query command not working with my index\n\nThe `/query` command only works with **integrated indexes** that use Pinecone's hosted embedding models. If you're using external embedding providers (OpenAI, HuggingFace, etc.), you'll need to use the MCP tools directly or wait for expanded support.\n\n### Assistant commands not working\n\nMake sure you have uv installed. uv is required for all assistant commands:\n\n```bash\n# Verify uv is installed\nuv --version\n\n# Install if missing\ncurl -LsSf https://astral.sh/uv/install.sh | sh  # macOS/Linux\n```\n\nAfter installing uv, restart your terminal.\n\n## Keywords\n\n`pinecone` · `semantic search` · `vector search` · `vector database` · `retrieval` · `RAG` · `agentic RAG` · `sparse search` · `document Q&A` · `citations` · `assistant` · `managed RAG`\n\n## License\n\nMIT License – see [LICENSE](./LICENSE) for details.\n\n\n**Have fun and enjoy developing with Pinecone!** 🌲\n",
  "bytes": 7242,
  "sha": "5bcdea76ce53de71f0c0adee1c9dfa7601a253a8f86469ea55b8113102820d37",
  "repo_slug": "pinecone-io/pinecone-claude-code-plugin",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_pinecone_io_pinecone_claude_code_plugin__dff7fd19/readme"
}