Search
@fragmentsx/vite-plugin
fireworksx
Vite plugin для consumer'ов Fragments. Автоматически генерирует `<script type="importmap">` и virtual shim modules для bare imports в externally-loaded ESM артефактах.
...more@fragmentsx/fragments-client
fireworksx
Клиентское ядро Fragments: загрузка и кеширование областей (area), фрагментов и css-чанков, метрики. Транспортный слой, на котором строится `@fragmentsx/runtime-react` — сам по себе фреймворк-агностичен и не зависит от React.
...more@fragmentsx/runtime-react
fireworksx
React-рантайм для встраивания опубликованных фрагментов Fragments в приложение потребителя. Строится поверх `@fragmentsx/fragments-client` — берёт готовый клиент и раскладывает его в провайдеры, хуки и компоненты React.
...moredocconvert-local
DocConvert Contributors
Clean Excel & Word → Markdown for RAG pipelines. Works 100% offline — no API keys, no cloud upload.
@megallm/router
priyanshu9588
Intelligent LLM prompt routing for MegaLLM — heuristic + optional ML-powered (MobileBERT zero-shot) model tier selection
@hartl-services/payload-s3-cache-storage
wolfi1616
Payload CMS plugin: S3 storage with local TTL-cached reads
@roadiehq/rag-ai-backend-embeddings-aws
roadiehq-david
The AWS (Bedrock) backend module for the @roadiehq/rag-ai plugin.
@roadiehq/rag-ai-backend-embeddings-openai
roadiehq-david
The OpenAI backend module for the @roadiehq/rag-ai plugin.
@roadiehq/rag-ai-backend-retrieval-augmenter
roadiehq-david
No description available
@roadiehq/rag-ai-node
roadiehq-david
No description available
@roadiehq/rag-ai-storage-pgvector
roadiehq-david
No description available
tempo9
Jiejing Zhang
Reserved for Tempo9 — server-grade LLM inference engine (Apple Silicon and CUDA). Real releases will appear here.
French Illustration Skill
yanliudesign
Hand-drawn editorial poster skill, prompts, and examples · 抽象线条海报生成 Skill
Csift
wdhwg001
The missing tool to sift your Claude Code sessions: regex search across all record types, recover files, extract images, inspect subagent topologies, match between plan files and sessions.
...moredsh-openai-account-connector
lxy2717
Official OpenAI account model connector for DeepSeek Harness
haiqv-ai
OpenAI-compatible LLM client with timeout and runaway protection for self-hosted models
hgraph-persistence
Durable record/replay and frame storage for hgraph
@buddhilive/bah-llm-buddhi
buddhilive
BuddhiAI Studio local inferencing adapter for the BuddhiAI Harness LLM seam
chunkr-rs
Blazingly fast document and text chunking library for LLMs and RAG in Rust and Python
@requestnetwork/ethereum-storage
request-bot
Request Network storage layer based on ethereum.