Six task recipes for everyday work

These recipes show common ways to get useful work done in Babbily. Treat each one as a starting point, then adjust the prompt, model, files, tools or plugins to fit your work.

Analyze a file

Use this when you want Babbily to summarize, compare, extract, rewrite or turn an attachment into action items.

1. Open a chat. Start a new one, or open the thread where the file belongs.

2. Attach the file. Any chat model accepts supported files. Add one or a few relevant files with the most useful material first, since long files can fill the context window.

3. Ask for the outcome. Name it: a summary, risks, a table, action items, a comparison, a rewrite or a plan.

4. Continue in the same thread while the file context is still useful.

“Summarize this document for an executive audience. Include open risks, decisions needed, and a five-item action list.”

Don’t upload secrets, private keys, payment details or files you aren’t allowed to use in Babbily.

Research with citations

Use this when a reply needs current information, public sources or a broader research report.

1. Choose the tool mode. Auto tool mode handles most research. Use Manual > Web search for a focused, current lookup, or Manual > Deep research for broader multi-source work.

2. Ask a source-aware question. Say what to research and which sources or time window matter. Add “show me photos of” when you want pictures of the subject with the answer.

3. Name the structure: a brief, a comparison, a timeline, pros and cons, a source table or a recommendation. Deep research writes the full report into Canvas.

4. Review citations. Open cited sources before you rely on important conclusions.

5. Check the report status. If the chat says coverage is partial, or Canvas labels the report incomplete, narrow your next request to the missing sections.

If you select Plugins for deep research, it reads from them but doesn’t send or change anything in connected apps, and it doesn’t pause for approval. Ask for any app action in a regular message once the report is ready.

Turn a chat into a document or slide outline

After a useful chat, brainstorm, research run or meeting-prep thread, reopen it and name the deliverable: a memo, brief, meeting notes, project plan, document outline or slide outline. If you use Google Docs, Google Slides, Notion or another supported app, connect it first and ask Babbily to prepare the content there when available. Review any draft, send, update or workspace action before you approve it.

“Turn this thread into a six-slide outline for a customer strategy review. Include slide titles, bullets, and speaker notes.”

Summarize email with a Plugin

Connect Gmail or Outlook in Settings > Plugins, then ask naturally: what to find, who or what to search for, and what you want back, such as open questions, decisions, blockers, next steps or a draft reply.

Approve actions carefully. Choose Allow on the approval card, or Send on an email draft. Typing “approved” in the chat doesn’t run anything, and cards expire after 24 hours.

“Use Gmail to find the latest thread with this customer, summarize open questions, blockers, and suggested next steps, and draft a reply.”

Create or edit an image or video

Every chat model can create images and videos with the built-in Image generation and Video generation tools. Ask for one in Auto tool mode, choose Manual > Image generation or Manual > Video generation, or type /create-image or /create-video.

• To edit an image, attach it or refer to one generated earlier in the chat. To animate an image or edit a video, attach it. Each request uses one source.

• Describe the subject, style, composition, motion and constraints, and ask for a square, portrait or landscape shape in your prompt.

• For a new video, set Video duration (3 to 10 seconds) and Video aspect ratio after you select Video generation. These settings don’t apply when you edit a video.

• Save, download or ask for another version.

Translate and continue in chat

Open Translate from the sidebar and choose your languages, or use auto-detect when available. Paste your text, or dictate it where your browser and device support voice input, and pick a style chip if you want the translation to sound different. When you want help replying or editing, choose Continue in chat to open a chat with the original and the translation as context.

“Rewrite this translated reply to sound polite, concise, and customer-ready.”

Read the full docs ↗