Remarc is a macOS feedback tool that captures context—text selections, screenshots with annotations, web elements, or voice recordings—and attaches it to comments for AI coding agents to read.
This means when you point out a problem to an agent, it gets the exact screenshot, selected text, or web data you saw, not just your description of it.
The app keeps everything local unless you explicitly share it with an agent, and integrates with Claude, Cursor, and other AI tools via plugins.
What happened
Remarc, a macOS app, lets users comment on text, screenshots, or web elements and attach those comments with full context to AI coding agents like Claude, Cursor, and others via integrations.
Why it matters
AI agents currently lose context when users describe feedback verbally or in separate prompts. Remarc preserves the original selection, screenshot, or web element data with each comment, so agents can act on what the user actually saw instead of a paraphrase—reducing back-and-forth and manual context rebuilding.
What to watch
The app runs locally on macOS 14.0 or later with no accounts or telemetry; data stays on your Mac unless you send it to an agent or export it. It is free and open-source under MIT license, with a companion Chrome extension for capturing web context.
Ask the AI about this article →
Remarc addresses a real friction point in AI-assisted coding: when a developer spots something that needs fixing, they must either describe it in text (losing visual context), take a screenshot and explain it separately, or copy-paste code fragments—all of which require the AI agent to reconstruct the developer's understanding from scratch. By bundling the original artifact (selected text, screenshot, web element) with the human feedback as a structured comment, Remarc lets agents work directly from what the developer saw, not from a lossy description.
The tool's design philosophy—local storage, no accounts, privacy by default—reflects a deliberate choice to make it a thin layer between the developer and their agent rather than a centralized service. The support for multiple input methods (text selection, screenshots with annotation tools, web element capture via Chrome extension, and voice transcription on macOS 26+) recognizes that different feedback situations call for different media. The integration with the Model Context Protocol (MCP) and agent plugin systems (Claude Code, Cursor, Codex, OMP) ensures that the context Remarc captures can be read by whatever agent the developer is using without lock-in.
For example, today's edition would include:
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · takes 30 seconds · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. Q&As are published on this page for other readers too.
Chinese large-model developer Z.ai says it can now support large-scale inference using roughly 100,000 domesti…
Analyst Ming-Chi Kuo says Nvidia has revived the Rubin CPX AI accelerator with a substantially redesigned arch…
A UK study by UK AI Security Institute and Limbic AI surveyed 6,474 British adults
Broadcom's Clayton Donley says companies are doing mission-critical work with AI agents quickly, but without t…
OpenAI released a new evaluation framework on July 17, 2026, urging companies to measure AI ROI by 'useful out…
As AI agents perform real business tasks, 'Agentic Identity' (giving each AI a unique employee-like ID) and 'D…
