Ai Research Python Packages

scitex

Umbrella package for SciTeX — reproducible science from raw data to manuscript

research-hub-pipeline

AI-operable research workspace for Zotero, Obsidian, and NotebookLM. Use any two, or all three, through CLI, MCP, REST, and dashboard.

scitex-dataset

Multi-domain scientific dataset fetcher (neuroscience, biology, medical)

airas

AIRAS - an open-source project for research automation

hn-tech-signal-mcp

MCP server for global tech & AI signal intelligence via HackerNews, arXiv, Lobste.rs and GitHub

swarms-cloud

Deploy your autonomous agents to production grade environments with 99% Uptime Guarantee, Infinite Scalability, and self-healing.

perplexity-agent-mcp

Zero-dependency, single-file MCP server for Perplexity's Agent API. Python stdlib only — no MCP SDK, no HTTP library, nothing to audit but one file.

autocrucible

Autonomous experiment platform — let LLM agents optimize anything through iterative edit-evaluate loops

notebooklm-skill

NotebookLM does the research, Claude writes the content. Research → Synthesis → Content Creation → Publishing. Claude Code Skill + MCP Server.

rlm-code

The Research Playground for the RLMSs and Coding Agents

llm-perplexity-agent

Zero-dependency, single-file MCP server for Perplexity's Agent API. Python stdlib only — no MCP SDK, no HTTP library, nothing to audit but one file.

inflion

Observability and influence tracing infrastructure for multi-agent AI systems.

llm-tokens-atlas

Open benchmark of LLM tokenization calibration across providers.

cogitator

A Python toolkit for chain-of-thought prompting 🐍

halftrace

Diagnostics for agent loops. Measure how fast each capability decays over long trajectories.

traceiq

Observability and influence tracing infrastructure for multi-agent AI systems.

autoevolve

Companion tools for Karpathy's autoresearch - smarter evaluation, guided steering, and multi-agent competitions for GPT pretraining

nouse

A persistent epistemic substrate for AI. Nous treats language models as larynx, not mind, and benchmarks epistemic structure instead of output fluency.

rigorgraph

Local-first claim-evidence graphs and deterministic audit reports for AI-assisted research.

nearlossless-context

Near-lossless long-context KV compression under fixed VRAM, critical-span retention, query-unknown discovery, and measured peak/quality Pareto.

zotero-keeper

Zotero 10+ Local API and PubMed MCP research workspace: Keeper 2.1, PubMed Search 0.6.1, and a one-click VS Code extension.

autojudge

Companion tools for Karpathy's autoresearch - smarter evaluation, guided steering, and multi-agent competitions for GPT pretraining

autosteer

Companion tools for Karpathy's autoresearch - smarter evaluation, guided steering, and multi-agent competitions for GPT pretraining

model-evolution

Model Evolution keeps the reasoning behind a machine-learning model connected to the exact data, code, runs, and artifacts that produced it.