Search

The Unoverse front door — create a Studio project, a universe, or a client app, and launch Studio.

Local visual editor for hand-coded HTML/CSS — captures edits as patches, Claude reconciles them into source

Deck TUI Ink per-progetto della famiglia loom: legge tasks.md e spawna sessioni Claude Code bound via LOOM_TASK

Original open-source agentic coding assistant CLI with persistent API keys, improved terminal TUI, slash commands, and NVIDIA NIM support.

Local coding-agent Skill inventory and source atlas

Layered verification pipeline for AI-agent-written code: cost-ordered gates, fail-fast, machine-readable verdicts for self-correction loops. Runs on Node, Bun and Deno.

Analyze images locally with Ollama instead of spending cloud multimodal tokens.

Servanda §3.4 extraction: tool-less, schema-bound, single-persona commitment extraction from signal envelopes

Server-side helper for minting Quintra AI Assistant widget sessions — builds, signs, and sends the HMAC-authenticated request to /widget/session.

The canonical, format-agnostic content and layout schemas shared by ooxml.js, odf.js, and documents.js -- pure Zod schemas, no behaviour.

Document ingestion and OCR: byte-level format detection, text-layer-first reading, local PP-OCR engines, an escalation policy, and the eval harness that settles the thresholds with evidence. The TypeScript half of one contract shared with a Python impleme

支持图片和多页 PDF 的 OpenAI 兼容 OCR 命令行工具

Complete React workflow for browser screenshots, screen recordings, annotation, trimming, preview, and form uploads.

Versioned learning module, assessment, entitlement, and reward contracts for the Plasius ecosystem

Cross-platform Agent Skill and deterministic CLI for persistent teaching, Web Lessons, diagrams, and channel artifacts.

Read-only TUI to watch Claude Code multi-agent workflows live

Multi-agent command center for AI coding agents — from your phone, tablet, or any browser.

Vercel AI SDK tools for the Fystash multi-agent sandbox (create, spawn, drive, exec, destroy).

Audit a repo's AI-agent configuration — config, rules, styles, environment — against the AGENTS.md standard, for any coding agent (Claude Code, Codex, Kimi, Cursor, Antigravity, Gemini).

The shared contract every Runbit package depends on. Almost never changes.