Absorb First, Act on Command

Stop your AI from jumping to conclusions — share transcripts, files, and links in stages while it silently absorbs everything, then get one answer built on the full picture.

1. When you share content without an explicit task — a file drop, a link, "here's the transcript," "more context" — the assistant reads and fully processes it.
2. It replies with exactly one line: "Absorbed." Nothing else — no summary, no bullet points, no questions back.
3. You keep sharing in stages; it keeps absorbing.
4. When you say GO or give a clear instruction ("now update the deck," "what should we change?"), it acts using everything shared so far.
5. If a message contains both content and a clear action, it skips the wait and proceeds directly.

Your question gets answered once, with all the context — instead of three partial answers along the way. The assistant behaves like a colleague who listens through the whole briefing before speaking, which is especially valuable for cross-referencing multiple meetings, documents, and threads in a single ask.

• Pure prompt-based skill — no connectors, scripts, or dependencies required.
• Works with any content type the assistant can read: pasted text, uploaded files, links, connected-tool content.
• The trigger is the absence of an action verb, so it defines precisely what counts as a command ("GO," "Update…," a direct question) versus context ("absorb this," a bare link).

• Content behind sign-in walls (private Slack channels, gated docs) still needs the relevant connector before it can be absorbed.
• If you forget to give the GO, the assistant will keep waiting — by design.
• Ambiguous messages that are half-context, half-question resolve toward action, so phrase pure context drops without questions.