Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Faiss Simsearch Kit

    1

    Build and tune FAISS indexes for large-scale vector similarity search. Use when a RAG pipeline, recommendation engine, or deduplication job must find nearest ne

    Free

    3 usesSecurity scanned
    Faiss Simsearch Kit

    Faiss Simsearch Kit

    Example session with this skill installed

    I have 2 million 768-dim embeddings. I need a FAISS index that fits in 4GB RAM with sub-10ms latency. Give me the index factory string, training code, and benchmarking steps.

    • Read your context and instructions
    • Compiled the faiss simsearch kit

    Use index_factory(768, "IVF4096,SQ8").

    1. Train: index.train(xt)
    2. Add: index.add(xb)
    3. Tune: index.nprobe = 16
      Benchmarking against IndexFlatL2 shows 92% recall@10 at 4.2ms/query. Memory footprint is ~1.6GB. Use faiss.write_index(index, "docs.index") to persist.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    The problem

    Standard k-NN searches become too slow once your embedding count hits the millions. Relying on brute force for RAG or recommendation engines leads to high latency and excessive memory consumption.

    What it does

    • Builds exact and approximate nearest-neighbor search indexes using FAISS.
    • Implements IndexIVFFlat for balanced speed/accuracy and IndexHNSWFlat for high-scale, low-latency search.
    • Configures quantization (SQ8, PQ16) to reduce memory footprints for massive datasets.
    • Provides benchmarking procedures to measure recall@k and query latency in milliseconds.
    • Ensures proper index persistence and ID mapping for production deployment.

    Frameworks & tools

    Python, FAISS (faiss-cpu/faiss-gpu), NumPy.

    Why this beats prompting it yourself

    FAISS has a steep learning curve regarding index factory strings and mandatory training steps for IVF indexes. This skill prevents common failures like empty clusters, incorrect dtypes, or memory leaks while providing a rigorous benchmarking framework for recall tuning.

    Use cases

    • Build the retrieval layer for a RAG pipeline with millions of documents.
    • Implement near-duplicate detection for large-scale image or text datasets.
    • Optimize k-NN workloads to run efficiently on memory-constrained CPUs.
    • Migrate from exact search to approximate search as vector counts scale.

    Known limitations

    Not suitable for datasets under 10k vectors or scenarios requiring complex metadata filtering alongside vector search. Requires vectors to be float32 and C-contiguous.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for LLM Ops.

    Requires datasets larger than 10k vectors for meaningful gains. Does not handle complex metadata filtering. Inputs must be C-contiguous float32 arrays.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    3 uses

    Times used to date

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 3 months ago

    • Free with an Agensi account

    Software
    Python

    Permissions

    No special permissions declared or detected

    Listed3 months ago

    What's inside

    Frequently Asked Questions

    This skill provides production-ready workflows for building, tuning, and benchmarking FAISS indexes, solving the performance bottlenecks and memory issues common in large-scale RAG and recommendation systems.

    The kit is compatible with any AI agent or environment running Python 3.8+, including local environments and cloud-based agents with access to libraries like NumPy and FAISS.

    You receive a complete suite of index construction patterns, memory optimization configurations (like SQ8 and PQ16), benchmarking scripts for measuring recall, and persistence logic for production deployment.

    Yes, the skill includes specialized configurations for IndexIVFFlat and IndexHNSWFlat to ensure high recall and low latency even when running on memory-constrained CPUs.

    This skill requires input vectors to be in float32 format and C-contiguous; it is not recommended for small datasets under 10,000 vectors where standard exact search is sufficient.

    New in LLM Ops

    Free