e2b-dev/awesome-ai-sdks

A database of SDKs, frameworks, libraries, and tools for creating, monitoring, debugging and deploying autonomous AI agents

What it is: This project is a curated list (an “awesome” list) of SDKs, frameworks, libraries, and tools used for creating, monitoring, debugging, and deploying autonomous AI agents. It’s a single place to discover many relevant resources.

What’s inside: The README lists 11 specific tools, each with a short description and direct links:

  • E2B – An “operating system” for AI agents: a set of tools, APIs, and cloud environments. Includes links to its web, GitHub, and social media.
  • AgentOps – Tools to make agents work, including graphs, monitoring, and replay analytics.
  • Chidori – A reactive runtime for building AI agents (supports Node.js, Python, Rust; currently in alpha).
  • Fixie – A platform for conversational AI, allowing agents in any language.
  • Helicone – An open‑source observability platform for GPT‑3 (tracks usage, cost, latency).
  • LangChain – A framework to simplify building applications with large language models.
  • Langfuse – Open‑source analytics for LLM apps (currently in closed alpha).
  • LangSmith – A unified platform for debugging, testing, evaluating, and monitoring LLM applications (closed beta).
  • SID – Data infrastructure for AI, letting developers connect to customer data via a single button and API.
  • Steamship – A platform for building, scaling, and monitoring AI agents with serverless hosting, vector search, webhooks, and more.
  • Vercel AI SDK – An open‑source library for building AI‑powered user interfaces.

Each entry appears in a collapsible section with links to the project’s website, GitHub, and sometimes Discord or Twitter.

Contribution: Users can add or improve items by submitting a pull request (the README explicitly invites this).

Who is behind it: The list is maintained by the team behind e2b (e2b‑dev). E2b describes itself as building an operating system for AI agents—a set of tools, environments, SDKs, and APIs that work with any tech stack.

Community and contact: The README provides links to Discord, Twitter (including a Twitter community), a scheduling link for calls, and an email (hello@e2b.dev). It also points to a companion “awesome‑ai‑agents” database.

Important caveat: The list is explicitly “according to our best knowledge, although definitely not comprehensive.” It’s a starting point, not an exhaustive directory.

Related

  • Project
    mem9-ai/mem9

    mem9 provides a persistent, shared memory layer for AI agents, allowing them to retain context across different sessions, machines, and platforms through a unified API.

  • Project
    mmalmi/nostr-vpn

    nostr-vpn is a Tailscale-style private mesh VPN using Nostr identities for enrollment and a FIPS-based data plane, with an experimental Bitcoin-paid exit-node marketplace. It offers a CLI, daemon, and native apps across desktop and mobile, plus server packages for StartOS and Umbrel. The fork claims ~3× throughput over the original FIPS implementation. Enrollment requires explicit admin approval; paid exits use Cashu channels with per-byte pricing.

  • Project
    AgenticAiLabs/Ai-Engineering-Roadmap

    A free, community-curated, and modular curriculum that guides self-taught learners through the entire path of becoming an AI engineer, from foundational mathematics to building autonomous agents.

  • Project
    traceroot-ai/traceroot

    TraceRoot is an open‑source platform that records AI agent calls, flags bad behaviour with custom detectors, clusters issues into signals, and runs automated evals so a coding‑agent can suggest and verify fixes, closing a self‑improvement loop. It offers a CLI, Python/TS SDKs, and integrations with major agent frameworks and model providers, and can be used via hosted cloud or self‑hosted Docker Compose.