Remote ArcREMOTE ARC
Understand videos with source timestamps
Compare release videos, explain demonstrations and find technical details. Start with captions, then retrieve only the audio or frames needed.
With Arc
Analyze the public video links I provide on my connected computer. Read Media Intelligence, retrieve metadata and captions first, and compare concrete features and technical differences with links to the original timestamps. Download only bounded audio or video sections when evidence is missing; inspect sampled frames for visual claims. Deliver a cited Markdown report and verified evidence files, distinguish source claims from inference, and disclose unavailable or unverified portions. Check current connections, permissions and dependencies first. Use real tools and continue from observations; claim only results actually observed or verified. Inspect uncertain external effects before any repeat. Report specific missing capabilities.
Arc requirements
Public individual video links, Python 3.11+, Node and an authorized writable workspace on a connected computer.
Arc boundaries
Retrieval and local transcription run on the device; selected evidence enters your AI's model context. Captions can be wrong and frames are samples. Protected or bot-blocked sources may be unavailable. No account cookies.
With Connect: example request
Analyze the public video links I provide on my connected computer. Read Media Intelligence, retrieve metadata and captions first, and compare concrete features and technical differences with links to the original timestamps. Download only bounded audio or video sections when evidence is missing; inspect sampled frames for visual claims. Deliver a cited Markdown report and verified evidence files, distinguish source claims from inference, and disclose unavailable or unverified portions.
With Connect: how the work proceeds
Retrieve captions and locate relevant original-video timestamps. Inspect missing audio/visual evidence and verify report citations.
Connect tools and permissions
read_skill · read_file · write_file · start_process. Public individual video links, Python 3.11+, Node and an authorized writable workspace on a connected computer.
What useful evidence looks like
A cited report, timed transcript or sampled frames, evidence hashes, covered intervals and stated limitations.
Know the boundary
Retrieval and local transcription run on the device; selected evidence enters your AI's model context. Captions can be wrong and frames are samples. Protected or bot-blocked sources may be unavailable. No account cookies.