Stop building your AI stack from 40 open tabs.

jucelioalencar 1 · yesterday

Legible persistent agent memory: records observable execution events, consolidates only validated knowledge and explains every relation — MCP server on PostgreSQL 17 with a web explorer.

iOfficeAI 30k · 3 days ago

Office suite built for AI agents: one binary, no Office install, creates, reads and edits Word, Excel and PowerPoint, and renders them to HTML/PNG so the agent can see what it made. Ships as a skill.

Tencent 1.3k · 3 days ago

Tencent WeChat's universal multimodal embedding family (2B/4B/9B): one vector space for text, images, video, visual documents and interleaved inputs, with Matryoshka dimensions from 64 to 4096.

ovg-project 1.3k · 14 days ago

Virtualized, elastic KV cache for LLM serving on shared GPUs: reserve virtual memory, back it with physical GPU memory only when used — vLLM and SGLang, with a memory-limit CLI, router and sleep mode.

vllm-project 167 · yesterday

Rust gateway that gives vLLM a stateful, OpenAI-compatible Responses API: server-side conversation state, server-side tool loops, SSE and WebSocket streaming, background runs — Codex-ready.

CopilotKit 4.3k · yesterday

CopilotKit's self-hosted AI coworkers: each gets its own computer — a real browser with logins, files, granted tools — with every action decided before it runs and recorded after. Any AG-UI agent.

ralforion 186 · yesterday

Browser ontology workbench for OWL and SKOS on Streamlit + rdflib: OWL-RL reasoning, 22 SKOS checks, SPARQL console, bulk edits, gist starters and a vis-network graph — no Java, no Protégé.

NVIDIA-NeMo 1k · 2 days ago

Molt: NVIDIA's agentic-first RL framework in ~9K lines — Ray for placement, vLLM for rollout, AutoModel + FSDP2 for training — fully async, multimodal, multi-turn, scaling to 1T-class MoE.

arcboxlabs 3k · 6 days ago

Open-source container and VM runtime for macOS in pure Rust: drop-in Docker engine, sub-100ms agent sandboxes (`abctl claude`), full Linux VMs and throwaway macOS guests on one daemon.

bingreeky 381 · 10 days ago

JIT-Agent: a meta-agent that writes a task-specific harness on the fly — memory, planning, action and capability modules emitted as code — and revises it from traces at test time.

zvec-ai 2.8k · 2 days ago

zg: ripgrep, BM25 and vector search behind one local-first CLI for humans and agents — index a workspace once, search code, docs and data by meaning, then verify with exact text or regex.

diegosouzapw 61.6k · yesterday

Free MIT AI gateway for coding CLIs: one endpoint over 352 providers (150+ free tiers) and 1,200+ models with quota-aware auto-fallback, RTK+Caveman token compression, MCP/A2A and a desktop app.

deeplethe 4.5k · yesterday

DeepLethe's open 'enterprise world model': one Rust binary plus Postgres running a bitemporal knowledge graph with ontology packs, cited hybrid search, an agent harness and MCP. Air-gap ready.

apache 4.8k · yesterday

Apache-incubating agent workspace where the append-only run log IS the runtime: desktop, TUI, CLI and Eval are thin clients of one host, and every harness benchmark ships with per-task records.

rahulnyk 4k · 22 days ago

Notebook recipe that turns any text corpus into a concept graph with a local Mistral 7B via Ollama — chunk, extract concepts and relations, add proximity edges — for Graph RAG and KG QA.

eunomia-bpf 671 · yesterday

System-level observability for AI agents via eBPF and TLS tracing: correlates prompts and model calls with the processes, files and network the agent actually touched — no SDK, no proxy.

profullstack 202 · yesterday

btop-grade terminal dashboards in TypeScript: a one-import API over a typed-array framebuffer with differential rendering, Braille graphs and truecolor — zero runtime deps, Bun/Node/Deno.

FailproofAI 2.1k · 2 days ago

Observability plus policy enforcement for agent harnesses: records every run across 12 harnesses (Claude Code, Codex, Cursor, Hermes…) and blocks dangerous tool calls with 39 built-in policies.

tt-a1i 49.4k · yesterday

Agent skill for architecture, workflow, sequence, data-flow and lifecycle diagrams: the agent emits typed JSON IR, Archify compiles it deterministically to interactive HTML/SVG with PNG/WebM export.

lupppig 6 · 22 days ago

YAML-defined ETL/ELT engine with a CLI and self-hosted control plane: Postgres/MySQL/Mongo/CSV/REST/PDF in, upserts with cursors, validation and quarantine, durable workers — AI transforms optional.

theswerd 738 · 1 months ago

shadcn/ui registry of React components that faithfully recreate the Claude Code, Codex and Grok terminal UIs — headers, tool calls, diffs, prompts — built from real tmux captures.

ulab-uiuc 2.8k · 17 days ago

UIUC's LLM routing library: 16+ trainable routers (KNN, MLP, matrix factorisation, Elo, graph, BERT) pick the best model per query, with xRouteBench and a train/serve CLI.

agentdock-ai 6 · 9 days ago

TypeScript library giving a backend one agent runtime: session-scoped runs with streaming, approvals and cancel, a tool registry, injectable stores and a factory over nine AI SDK providers.

PunithVT 470 · 1 months ago

Self-hosted digital human: a photo plus 10s of voice becomes a real-time talking head — Whisper to LLM to Chatterbox TTS to MuseTalk lip-sync, streamed over WebSocket, with barge-in.

ashuttl 358 · 4 days ago

Six dependency-free terminal apps for weather, sun, moon, tides, radar and maps, drawn from free public data with no accounts or API keys — mouse-friendly TUIs that try to match your terminal theme.

bhimrazy 671 · 1 months ago

Receipt-to-JSON in one pip install: CLI, Python API and FastAPI service that send a receipt image to any OpenAI-compatible model and return merchant, totals and line items, plus a Tesseract module.

gridex 1.4k · 3 days ago

Native database IDE for Postgres, MySQL, SQLite, Redis, MongoDB, SQL Server and ClickHouse, with a built-in MCP server — 13 tools, 3-tier permissions, audit trail — and schema-aware AI chat.

Rath-Team 1.1k · 1 months ago

PyTorch-shaped multi-agent framework: Session is the tensor, Agent the layer, Workflow the module — plus a v2 durable runtime with checkpoints, leases, an effect ledger and human interrupts.

RyanCodrai 16.7k · 16 days ago

Rust vector index with Python bindings built on Google's TurboQuant: no training step, online ingest, hand-written SIMD kernels — 10M x 1536 vectors in ~4 GB, with allowlist-filtered search.

NVIDIA 3.7k · 3 days ago

NVIDIA's model-compression library: quantization (PTQ/QAT, FP8/NVFP4), pruning, distillation, NAS and speculative decoding over HF/PyTorch/ONNX, exported to TensorRT-LLM, vLLM and SGLang.

kelviq 176 · 9 days ago

Claude Code skill that reads your local session logs and answers 'where did my tokens go' in plain English: deduplicated totals, cost charged to the tool that caused it, 5-hour window state.

Untrivial-ai 10.9k · 3 days ago

Desktop agent IDE for fleets of coding agents: each task gets a worker with its own workspace and agent CLI, while a project orchestrator plans work and handles CI fixes, conflicts and reviews.

wlsdks 93 · 3 days ago

Desktop app plus MCP server keeping a codebase's ontology as reviewable Markdown in the repo: domains and capabilities linked to code evidence, compiled to a graph for blast-radius queries.

mtrnix 97 · 3 days ago

Self-hosted agent memory stack in Docker: Postgres + Qdrant + Neo4j hybrid retrieval, a temporal knowledge graph, ontology layer and freshness checks behind one MCP-native API.

mikehasa 692 · 3 days ago

Local-first work receipts for coding agents: reads the session logs Claude Code, Codex, OpenCode and Hermes already write and reports what each task did, what it cost, and whether a check proved it.

alexzhang13 198 · 13 days ago

Speculative programmatic tool calling: while the model is still streaming a code block, the harness launches the tool and sub-LLM calls it is about to make, so blocking calls overlap generation.