Skip to main content
JobPlugins

Best Claude plugins for AI engineers

Tracing, vector search and multimodal apps.Our top 3 picks, ranked. Start with number 1.

What plugins can do for you

AI engineers need to see what their apps are doing. The plugins here trace and debug LLM apps with Langfuse, tune vector search with Qdrant, and build multimodal apps with Pixeltable. All three are verified, and they are aimed at developers who already work with these tools.

  1. #1Anthropic verified

    Langfuse

    Saves you time tracing and debugging LLM apps.

    Skills for working with Langfuse, the open-source LLM engineering platform for tracing, prompt management, and evaluation.

    View details — Langfuse

    Shows you how, in 4 steps

    How to add Langfuse to Claude

    Follow these steps once. It takes about a minute.

    1. Open Claude on claude.ai or in the Claude desktop app, and sign in.
    2. Go to Customize, then Plugins, then Discover.
    3. Search for “Langfuse” and select Add.
    4. If the plugin lists connected services, open its Connectors tab and connect them.

    Plugins are available on paid Claude plans. At work, your administrator may need to approve them first.

    73.5K installs
  2. #2Anthropic verifiedBy Qdrant

    Qdrant

    Saves you time tuning vector search.

    Agent skills for Qdrant vector search: scaling, performance optimization, search quality, monitoring, deployment, model migration, version upgrades…

    View details — Qdrant

    Shows you how, in 4 steps

    How to add Qdrant to Claude

    Follow these steps once. It takes about a minute.

    1. Open Claude on claude.ai or in the Claude desktop app, and sign in.
    2. Go to Customize, then Plugins, then Discover.
    3. Search for “Qdrant” and select Add.
    4. If the plugin lists connected services, open its Connectors tab and connect them.

    Plugins are available on paid Claude plans. At work, your administrator may need to approve them first.

    66.7K installs
  3. #3Anthropic verified

    Pixeltable

    Saves you time building multimodal AI apps.

    Build multimodal AI applications with Pixeltable -- tables, computed columns, embedding search, UDFs, tool-calling agents, and 25+ AI provider integrations.

    View details — Pixeltable

    Shows you how, in 4 steps

    How to add Pixeltable to Claude

    Follow these steps once. It takes about a minute.

    1. Open Claude on claude.ai or in the Claude desktop app, and sign in.
    2. Go to Customize, then Plugins, then Discover.
    3. Search for “Pixeltable” and select Add.
    4. If the plugin lists connected services, open its Connectors tab and connect them.

    Plugins are available on paid Claude plans. At work, your administrator may need to approve them first.

    62.1K installs

30 more plugins for this job

Not ranked by us. Sorted by how many people installed them.

  • Omniout: LLM Retries and Model Fallback

    Make your app's AI calls reliable: timeouts, retries with backoff, rate limit handling and fallback across OpenAI, Anthropic, Google and other models.

    Not verified21.1K installs
  • agent-memory

    Save 60-90% on LLM token costs with intelligent memory compression for multi-agent systems.

    Not verified12.5K installs
  • Auto Research

    Autonomous skill improvement through parallel research agents using the autoresearch methodology.

    Not verified8.1K installs
  • agent-architecture-designer

    60 specialized subagents and 62 team-orchestration presets (research, build, QA, Five Minds debate…

    Not verified3.8K installs
  • Agent Ingest, Audit & Optimize

    Audit external material and turn verified findings into safe, reversible agent improvements.

    Not verified3K installs
  • Prompt Optimizer

    Turn a rough draft into a clear, ready-to-send prompt for any AI.

    Not verified2.4K installs
  • Jev

    Typed decisions as a noun: adds $.jev in the engine.create fold, so any plugin asks Jev a map of Choice…

    Not verified2.2K installs
  • Ruflo Ai Team

    Give Claude a tenant-isolated AI team that divides work, coordinates progress, recalls approved context, and returns evidence-backed results.

    Not verified2K installs
  • OpenJev for Claude Code

    Jev System One for Claude Code — typed Choice/Noul/Score decisions via live TypeSafe API (same 5 tools as opencode-openjev, no mock).

    Not verified1.7K installs
  • together-ai-skills

    Agent Skills for Together AI platform — inference, training, embeddings, audio, video, images, function calling, and infrastructure.

    Verified1.5K installs
  • Agentic Atlas

    Do design work with Agentic Atlas over hosted MCP at agentic-atlas.dev/mcp/.

    Not verified1.4K installs
  • Configurable Council

    Convenes a council of AI advisors you configure, from a bundled preset or a lineup you design…

    Not verified1.3K installs
  • Build Harness

    Guides a person interactively through building their own multi-agent harness for their field in Claude Code, covering Context Engineering, Roles (subagents)…

    Not verified1.1K installs
  • Prompt Tightener: Clearer, Shorter LLM Prompts

    Review and rewrite prompts and system prompts for Claude and other LLMs: find repetition, filler and vague instructions, keep template variables…

    Not verified1.1K installs
  • ops

    Turns a pile of installed skills into a plan that picks which to run and in what order, runs them with backups and check-ins…

    Not verified1.1K installs
  • consensus

    Get unstuck by querying multiple AI providers (GPT-5.5, Gemini, DeepSeek) for diverse perspectives.

    Not verified1.1K installs
  • Boardroom

    Convene two or more of your Claude agents in one meeting with you.

    Not verified892 installs
  • Adaption

    Connect Claude to Adaption for dataset management and fine-tuning workflows.

    Not verified873 installs
  • emem

    Verifiable shared memory for AI agents: resolve a place to one canonical address, recall Ed25519-signed Earth-observation facts…

    Not verified873 installs
  • redis-development

    Redis development best practices — data structures, query engine, vector search, caching, and performance optimization.

    Verified793 installs
  • llm-secure-patterns

    Secure design patterns for Claude Code LLM applications: external content ingestion, endpoint hardening, output validation, system prompt design…

    Not verified763 installs
  • fleet

    Your claude agents list in a side pane, with optional cones columns.

    Not verified757 installs
  • CloudTalk VoiceAgent Builder

    Build, fix, validate, and test CloudTalk VoiceAgents.

    Not verified744 installs
  • PromptGuard

    LLM security for AI coding agents -- protect applications from prompt injection, PII leakage, and data exfiltration.

    Not verified744 installs
Show the other 6
  • spiceai/skills

    Skills for working with the Spice.ai runtime — data federation, acceleration, search, AI/LLM, and cloud management.

    Not verified729 installs
  • pinecone

    Pinecone vector database integration.

    Verified688 installs
  • EvalDossier

    Offline verification and fixed synthetic conformance for portable signed EvalDossier dossiers.

    Not verified641 installs
  • Skilldb

    Find and compare reusable AI agent skills using public catalog metadata and short previews.

    Not verified580 installs
  • Brainbase MCP

    Build, test, deploy, and monitor Brainbase managed agents safely.

    Not verified574 installs
  • validate

    Validation subagent that stress-tests plans, recommendations and architecture before you act on them.

    Not verified554 installs

Popular topics for this job

Quick answers

What is the best Claude plugin for AI engineers?
Our top pick is Langfuse. Saves you time tracing and debugging LLM apps.
How many Claude plugins are there for AI engineers?
We list 33 plugins for this job, 6 of them with Anthropic's verified badge. The top 3 are ranked by us, and the rest are sorted by installs.

New to plugins?

Adding one takes about a minute, and you can remove it any time.

How to install Claude plugins