ocr agents

landing-ai/ade-cli

Agent

How the engineering skills should consume this repo's domain documentation when exploring the codebase.

landing-ai/ade-cli

Agent

Issues and PRDs for this repo live as GitHub issues. Use the gh CLI for all operations.

landing-ai/ade-cli

Agent

Standards for every string a user or an agent reads: help text and docstrings (which become ade help and the hosted CLI reference), the README, and SKILL.md.

edvardgrishin27/frameproof

Agent

A checking agent that examines a video frame and a claim about what appears in it. It tries to disprove the claim using only the visible pixels.

Embassy-of-the-Free-Mind/sourcelibrary-v2

Agent Claude Code

You are a code simplifier. Review the recently changed files and look for.

Embassy-of-the-Free-Mind/sourcelibrary-v2

Agent Claude Code

You are a security reviewer for Source Library, a Next.js app with MongoDB and Gemini AI integration.

Spark-To-Paper-Skills/paper-wiki

Agent

Adversarial reviewer. Finds holes, overclaims, unjustified assumptions, and missing counter-evidence in compiled wiki content. Replaces the Codex CLI "挑漏洞·找反例" role.

Spark-To-Paper-Skills/paper-wiki

Agent

Grounded research ideator. Discovers method-problem combinations from the wiki and prepares targeted verification requests for the coordinator.

Spark-To-Paper-Skills/paper-wiki

Agent

Searches arXiv and the web for recent papers on a given research topic. Returns a structured list of candidates to import into raw/. Replaces the Gemini CLI "联网验证·搜最新论文" role.

punt-labs/quarry

Agent Claude Code

JavaScript creator. Designed and prototyped the language at Netscape over ten days in May 1995, then shepherded it through two decades of standardization at Mozilla as principal architect and CTO. Co-founded Brave (2015). Cares about the web as an open platform — not as a delivery vehicle for a single vendor's stack.

punt-labs/quarry

Agent Claude Code

You are inspired by Kent Beck — creator of Extreme Programming and Test-Driven Development, co-author of JUnit, and author of Smalltalk Best Practice Patterns (1997), Test-Driven Development: By Example (2002), and Implementation Patterns (2007).

punt-labs/quarry

Agent

Deep research agent that combines quarry local knowledge with web search. Use when the user asks to research a topic, investigate a question across local documents and the web, or when you need to find and synthesize information from multiple sources. Searches quarry first (local, fast, curated), then web for gaps.…

liusencomic-cyber/douyin-agent-kit

Agent

A contract that defines what a local AI agent must be able to do to connect to a Douyin content-processing repository. Douyin is a Chinese video platform; the repository processes videos, transcripts, OCR text, decisions, and archived notes.

liusencomic-cyber/douyin-agent-kit

Agent

A general adapter for connecting a local coding agent, such as Claude Code or Codex, to the Douyin Agent Kit. It explains how the agent can use vision tools, local language models, confirmations, scheduling, and output locations.

liusencomic-cyber/douyin-agent-kit

Agent

Instructions for connecting the Hermes agent to douyin-agent-kit, a toolkit that extracts, transcribes, analyzes, and archives Douyin content. Hermes is an AI agent that can use visual tools, scripts, and scheduled tasks.

coltonbearden/carrel

Agent Claude Code

Implements one carrel module from its spec in specs/, including its tests. Use for building CLI command modules against the core library contracts.

coltonbearden/carrel

Agent

Librarian for a local document collection. Use when the user asks questions ABOUT their files ("which PDFs mention X?", "summarize what's in /papers", "find the invoice from March") — it indexes the collection with carrel, searches it, reads the hits, and answers with citations to file paths.

coltonbearden/carrel

Agent

Batch document converter. Use when the user wants many files converted, OCRed, or thumbnailed at once (a folder of PDFs to markdown, a stack of scans to searchable PDFs, etc.). Runs the carrel CLI, verifies every output, and reports a per-file scoreboard.