Skip navigation
this is where GPT-6 Astra is getting f...cking illegal
0:00 / 0:22

this is where GPT-6 Astra is getting f...cking illegal

Lummox

@Lummox_eth8.2K followers

Visit on X
498 likesDownloadOriginal post

43,873 views

this is where GPT-6 Astra is getting f...cking illegal 10 open-source GitHub repos I'd use to build the system around the model 01 PydanticAI ▸ → typed agents + structured outputs 02 LangGraph ▸ → state + long-running agent workflows BUILD THE AGENTS 03 Agno ▸ → agents + teams + workflows 04 smolagents ▸ → agents that execute through code GIVE THEM MEMORY 05 Mem0 ▸ → persistent agent memory 06 Graphiti ▸ → temporal knowledge graphs GIVE THEM A RUNTIME 07 E2B ▸ → isolated sandboxes for AI-generated code 08 Daytona ▸ → infrastructure for running agent workloads TEST EVERYTHING 09 DeepEval ▸ → evals for LLM systems 10 Langfuse ▸ → traces + evals + production monitoring the flow: request → agent → memory → tools → runtime → eval → retry → deploy that's the part people underestimate the model is one folder the system around it is the product ⭣

25 Comments

Comments from the original post on X (Twitter)

  • Grootex's profile picture

    @Grootexbt

    GitHub allows you to create everything what you can imagine Crazy staff

    1 like

  • Ridark's profile picture

    @ridark_eth

    The command flow in this terminal looks smooth

    1 like

  • Lummox's profile picture

    @Lummox_eth

    Glad to hear it

  • Mura's profile picture

    @0x_mura

    that's real cheat bro

    1 like

  • Hussain Hashim | Building SundayBack's profile picture

    @itsthedonhashim

    @Lummox_eth these are solid picks! something i've done is set a checklist for testing state updates. catching those early saves headaches later.

    1 like

  • Gipp 🦅's profile picture

    @gippp69

    the flow fits perfectly here

    1 like

  • Lummox's profile picture

    @Lummox_eth

    Have the same feels Gip

  • Umar Saeed's profile picture

    @UmarSaeedX

    using these together could lead to complex integrations, but that's part of the challenge.

    1 like

  • unfair.so's profile picture

    @unfairdotso

    traces should feed the training loop, not just a dashboard

    1 like

  • monokern's profile picture

    @monokern

    alpha here, thx for sharing

    1 like

  • beamnxw ./'s profile picture

    @beamnxw

    building around the model is where the real work starts

  • ALEXYZ's profile picture

    @Alexvx_nft

    pydanticai plus langgraph clean foundation for agents

  • Robin | Poker x AI's profile picture

    @RobinHill85

    The repo list has agents, memory, and runtime covered. But making that eval loop work smoothly as a feedback cycle? That’s the true test.

  • PolyScripts | Prediction Market Intelligence's profile picture

    @polyscripts

    Robust pipelines scale autonomous agents.

  • Rali0s's profile picture

    @3xCypher

    Holding onto to this. +1

  • 安叫兽|Bird🕊️ 🔶 BNB's profile picture

    @ajs6888

    PydanticAI 和 LangGraph 放一起,刚好补两块短板

  • :)wsab's profile picture

    @b_kjomp

    how you have this interface ? give me pleaseee

  • Karpacho's profile picture

    @Karpachooa

    Another alpha group of GitHub repos Thanks a lot broski

  • Dekos's profile picture

    @PolyDekos

    Is it even legal for her to do that?

  • ShadowAguy's profile picture

    @shadowaguy

    ten repos later, astra is still mostly a ui with strong opinions about your folder structure.