Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

AgentDB Vector Search

Teaches Claude to build semantic search and RAG pipelines on top of the AgentDB vector database.

Dev & CodingIntermediate★ 371⑂ 36AI score 8/10Last updated: Jul 28, 2026

What it does

Gives Claude a concrete playbook for implementing semantic vector search with AgentDB.

  • CLI workflow: npx agentdb init/query/export/import/stats, with dimension, preset and in-memory flags
  • TypeScript API for storing embeddings, similarity search, and hybrid search (vector + metadata filters)
  • Guidance on distance metrics (cosine, euclidean, dot product)
  • Quantization options (binary 32x, scalar 4x, product 8–16x memory reduction)
  • Tuning notes on HNSW indexing, caching, and MMR for diverse results
  • MCP server setup so Claude Code can query the DB directly
  • Troubleshooting for slow search, high memory, poor relevance, dimension mismatch

Who it's for

  • Node.js/TypeScript developers building RAG chatbots or internal document search
  • Teams wanting a local, lightweight vector store instead of a hosted service
  • Anyone who keeps second-guessing embedding dimensions, similarity thresholds, and compression settings

Example uses

  1. "Index 500 internal wiki markdown files into a searchable vector DB" → init at dimension 768, then generate a batch-insert script
  2. "Write a RAG function that retrieves the top 5 chunks and injects them into the prompt" → code with threshold 0.7 and MMR enabled
  3. "My vector DB uses too much RAM" → switch to binary quantization and verify with the stats command

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/spencermarx/open-code-review/HEAD/.claude/skills/agentdb-vector-search/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the .claude/skills/agentdb-vector-search folder from the GitHub repo spencermarx/open-code-review into my ~/.claude/skills/agentdb-vector-search/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/spencermarx/open-code-review.git /tmp/ocr && mkdir -p ~/.claude/skills && cp -r /tmp/ocr/.claude/skills/agentdb-vector-search ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal and make sure Claude Code is installed.
  2. Clone the repository into a temp folder: git clone https://github.com/spencermarx/open-code-review.git /tmp/ocr
  3. Create the skills directory: mkdir -p ~/.claude/skills
  4. Copy the skill: cp -r /tmp/ocr/.claude/skills/agentdb-vector-search ~/.claude/skills/
  5. Verify Node.js 18 or newer: node -v
  6. Export an embedding API key: export OPENAI_API_KEY=sk-... (or plan to use a local embedding model)
  7. Restart Claude Code and try a prompt like "Set up an AgentDB vector store for semantic search" to confirm the skill triggers.
  8. Optional: register the MCP server with claude mcp add agentdb npx agentdb@latest mcp
View source on GitHub ↗License: Apache-2.0