Search

Valkey vector database adapter for cognee

Qdrant vector database adapter for cognee

OpenSearch vector database adapter for cognee

A fused Triton SwiGLU MLP kernel (gate_proj + up_proj + SiLU fused, TMA-based) for training on Hopper/Blackwell GPUs.

A python phishing script for login phishing, image phishing, video phishing and many more

Automatic provider/model fallback chains for DeepSeek Harness agents when LLM requests keep failing (retry exhausted, auth, quota, rate limit)

File storage utilities to store and retrieve files in a holochain DHT

Typed localStorage with initial values, groups and cross-tab subscriptions. SSR-safe, dependency-free, a few hundred bytes.

Local-first skill routing for agent builders — index SKILL.md bundles, get ranked skill plans with confidence and evidence, and serve them to any agent over MCP.

生成面试题+考察维度+A级参考答案的完整面试准备系统 | Interview Question Generator with Assessment Dimensions and Reference Answers

The first vision plugin for DeepSeek Harness, and the vision bridge for every text-only coding agent. Paste an image, get structured JSON evidence (OCR, layout, semantics).

Turn any technical book PDF into a Claude Code skill — ready to study, reference, and use while you work.

Agent-driven proactive memory CLI for AI agents — autonomously recall, maintain, and evolve persistent, source-grounded knowledge across sessions.

Convert documents (PDF, DOCX, XLSX, HTML, etc.) into Token-Dense Format (TDF), a lossless plain-text format that cuts LLM token costs by 28-47% compared to Markdown.

On-The-Fly Weight Synthesizer (OTF-LLM Engine) for ultra-fast, low-VRAM LLM inference with Fused Triton INT4 kernels.

Boruna: a deterministic, capability-safe programming language and framework for LLM-native applications