Search
@kamanai/ui-chatbot
yspl
Kaman's chat thread for React — messages, streaming, tool calls and citations.
@benrong/dsh-prompt-polish
benrong
Prompt-polish button beside the composer model seat for DSH, porting linshenkx/prompt-optimizer's optimizationMode axis, mustache template engine, template registry and variable extraction.
...more@jeffreyjyz/opencode-context
jeffreyjyz
Context-window breakdown for opencode: a /context dialog plus a context_breakdown tool showing where the window went (system prompt, tool definitions, thinking, tool I/O, messages), delta.app style.
...moredsh-prompt-zh
GitHub Actions
把 DSH 内置的系统提示词段落、MCP 指导说明与运行时注入消息本地化为简体中文(注入点 system-prompt/assemble 与 agent/pre-step)。工具的 description 与参数 schema 有意保持英文。
...more@drive9/drive9
qiffang
TypeScript SDK for drive9 — an agent-native filesystem with semantic search
dsh-model-think-level
gaohu9712
DSH 的第三方模型补齐思考等级:在官方 Models 页每个模型行注入 Thinking 开关和强度/模态编辑器,在输入框模型菜单加供应商→模型选择器、视觉徽章与等级行,并带搜索、拖拽排序、供应商启用开关;以及按路由接管请求头与 user-agent。
...moreamtal-inference
amtal
Amtal inference SDK for JavaScript and TypeScript.
@cuzom/eslint-config
GitHub Actions
ESLint flat config: recommended JavaScript and TypeScript rules, light typing, formatting left to Prettier.
@doc-cheap/ai-sdk
GitHub Actions
doc.cheap tools for the AI SDK – read passports, ID cards and driving licences into structured fields. Free: 100 documents every month, then $0.01 each
...more@tubeagentkit/ai-sdk
tubeagentkit
YouTube transcript tools for the Vercel AI SDK and Mastra: get YouTube video transcripts with timestamps, search YouTube, and list channel videos, via the GetYouTubeTranscript API.
...more@ctxwise/ai-sdk-liteparse
uditkumar01
Vercel AI SDK middleware that turns chat attachments (PDF, Office, images, scans) into LLM-ready text with LiteParse, in-process with no server, using vision only where OCR is not confident
...more@slices.sh/transcription.browser
smarcombes
slices v2: the transcription service, over the browser's own speech recognition. No key, no route.
@matteogiaquinto/speech-to-text
saveyh
Local speech-to-text for web apps. Add private Whisper-powered dictation to any input with no backend or API key.
@kubohiroya/turbowarp-kamishibai-app
kubohiroya
TurboWarp-Kamishibai — an AI-powered interactive storytelling system built on the TurboWarp ecosystem, using Teachable Machine recognition as story input
...more@tychenjiajun/pi-voice
tychenjiajun
Speech-to-text for Pi (local model or OpenAI-compatible remote ASR, compaction-style post-processing)
@veriworkly/ats-engine
gautam-raj
ATS resume checker and parser: reads a resume (PDF, DOCX, text, JSON Resume) the way an applicant tracking system does, scores it with a published deterministic rubric, and matches it to a job description. Browser- and edge-safe.
...more@akashpriyadarshii/jev-seo
akashpriyadarshii
Free Rust SEO and GEO toolkit powered by TypeSafe Jev: 64-rule audits, site crawls, AI citation checks, rank drift, CI gates, 19-tool MCP. Zero cost.
...more@horizonpay/invoice-ai
GitHub Actions
Official TypeScript SDK for the Invoice-AI API.
@caspianmd/caspian
GitHub Actions
Caspian in your terminal — a copilot on top of the AI coding CLIs you already have.
@qball-inc/elements
GitHub Actions
QBall Design System framework-free elements — plain TypeScript/DOM trust atoms, chat and knowledge-graph components that paint the @qball-inc/tokens CSS. No framework runtime.
...more