Everything in Aleria starts in chat. Type what you're trying to do, in plain language, and Aleria works out the steps. You don't choose a mode, pick an agent, or fill in a template: it's one conversation, and the shape of the answer depends on what you asked for.
Just ask
Type what you want. Aleria answers. That is the default experience, and it covers most daily work: questions, summaries, drafts, calculations, translations, code snippets, and quick data lookups.
You do not need to phrase things carefully. All of these work:
- "Rewrite this in a calmer tone."
- "What is the total across these three files?"
- "Make it look like a bank statement."
If Aleria misunderstands, correct it. The conversation keeps going.
Voice input
You do not have to type. Press the microphone next to the input, start talking, and Aleria transcribes as you go. This is useful when you are on the move, thinking out loud, or describing something too long to type, such as an incident report or a meeting takeaway.
Voice input supports English, Arabic, and many other languages out of the box. You can switch between typing and speaking mid-message.
If you want to capture a whole meeting rather than dictate a single message, use the Record option instead. See Record audio in chat.
The enhance button
Next to the send button is an Enhance button. If you are not sure your message is clear, press it. Aleria rewrites your draft into a sharper version that is more specific, better structured, and more likely to produce the result you want. Review the suggested wording, then accept it, edit it, or keep your own.
The enhance button is especially useful when:
- You are about to ask for a big deliverable, such as a deck, a report, or an app, and want to make sure the request is complete.
- You are new to Aleria and still figuring out how much to say.
- You want to turn a scribbled note into a well-formed task in one click.
Enhance never sends on your behalf. It rewrites the draft in place. Nothing is sent until you press send, so you always see the final wording before the chat runs.
Upload anything
Drag files straight into the chat or paste them from your clipboard. You can attach as many files as you want in a single message. Aleria reads them together, as one context.
What you can upload
- Documents: PDFs, including scanned ones, Word, Markdown, plain text, RTF, HTML, ebooks
- Spreadsheets: Excel, CSV, TSV, including formulas and multiple tabs
- Slides: PowerPoint and Keynote exports
- Images: photos, screenshots, charts, diagrams, scans
- Audio: meetings, voice notes, calls (transcribed on upload)
- Video: transcripts plus keyframe analysis
- Archives: ZIP files are unpacked automatically
- Code and logs: any text source file or log format
Aleria can handle very large files, such as full annual reports, whole codebases, or hours of audio. If something is too big, it will tell you and suggest splitting. This is rare.
Generate any kind of file
Ask for a deliverable, and that is what shows up, not just a description of one. Examples include:
- A PDF report with your logo
- A CSV of reconciled rows
- A formatted Word contract
- A slide deck with speaker notes
- An image
- Source code
- A ZIP of many files
Everything generated is downloadable, editable, and saved to the chat thread. Ask for revisions in plain language, such as "make the margins tighter," "use our brand colors," or "drop the footnotes." A new version replaces the old one.
Build and deploy apps
You can ask Aleria to build something, such as a dashboard, a calculator, a custom viewer, or an internal tool, and it does not stop at showing you code. It builds, tests, and deploys the app on a temporary Aleria address, such as yourteam.aleria.app/pipeline. You get a live link in chat. Share it, open it on your phone, or send it to a stakeholder before a meeting. See Apps for the full picture.
Examples of apps you can build:
- A one-page dashboard reading from your data
- A branded calculator or quote tool
- An interactive document viewer or data explorer
- A lightweight internal tool for approvals, form intake, or tagging
- A prototype to show stakeholders before committing to a real build
Choose a model and turn on deeper thinking
Most of the time you can let Aleria pick the right model for the job. If you want control, use the model selector near the chat input to choose a faster model for quick drafts or a stronger one for complex analysis. You can also turn on a thinking toggle so the model reasons more carefully before it answers, which helps on tricky tasks. See Choose a model.
When a chat becomes an agent
Some tasks do not finish in a few seconds. Examples include reviewing every lease in a 400-file data room, ingesting a database and building a schema map, or watching a folder for changes and summarizing what is new each morning.
When you ask for work like that, the chat promotes itself into a background agent. You will see a status panel with the steps it is running, the files it is touching, and a progress bar. You can close the tab. It keeps going.
When the agent finishes, or hits something it needs you for, you get a notification. Open the chat again and you will see a complete log of every decision it made and every action it took.
There is no "agent mode" button. The chat decides based on the task. You describe the outcome, and it figures out the shape of the work. See Agents.
What the chat has access to
Out of the box, the chat can reach:
- Your attached files and everything in the workspace library
- The workspace's sources, including databases you created inside Aleria, external systems you connected, and files your team uploaded
- Skills, both the built-in ones and the ones your team created
- The web, when the answer is not internal
- Your connected tools, such as email, calendar, and Slack, through Connectors
Anything the chat produces, such as a cleaned table, a generated report, or a query result, can be saved back as a source for your team to reuse.
Inline citations and sources
When Aleria gives you a factual claim, such as a number, date, or quote, it shows where the information came from. Each claim is linked to its source right inside the message.
How it works
- Click a citation marker next to a claim.
- A card opens showing the exact source: a file, a database row, a web page, or a connected system.
- You can open the source directly to verify it yourself.
For example, if Aleria says "Q2 revenue was $4.2M," the number will have a citation. Click it and you will see the spreadsheet or report the number came from.
Why this matters
Every message shows the references Aleria used to build the answer. This is useful when:
- You need to double-check a figure before a board meeting.
- You want to see which documents Aleria actually read.
- You are reviewing work for compliance and need a clear audit trail.
If a source looks wrong, tell Aleria which file or table to use instead.
Listen to a response
Every response has a Read aloud control. Press it and Aleria reads the message out loud, which is handy when you are multitasking or want to proofread by ear. Press it again to stop. See Listen to responses.
Branch from any message
You can branch a conversation from any message. This creates a full copy of the chat up to that point, so you can explore a different direction without changing the original thread.
When to branch
- You asked for a summary and now want to try a different format. Branch from the original request and ask again.
- A teammate wants to take the same context in a different direction.
- You want to test a risky edit or a bold idea without losing the good version you already have.
How to branch
- Open the menu on the message you want to branch from.
- Choose Fork from here.
- A new chat opens with the full conversation copied up to that message.
- Continue the new chat in whatever direction you want.
The original thread stays untouched. The branched chat has its own link and can be shared separately.
Sharing and continuity
Chats are scoped to your workspace, so teammates can pick up where you left off. You can share a thread with a direct link, branch a chat to take it in a new direction, or turn a recurring request, such as "every Monday, prep last week's pipeline summary," into a Workflow.
Inline shortcuts in the composer
Inside the message box, three single-character triggers bring up inline pickers so you can add context without leaving the keyboard:
@: attach a file or source from your workspace/: use a skill$: bring in an agent
Two app-wide shortcuts are worth knowing:
Cmd + K(Windows and Linux:Ctrl + K): search and jump to any chat, file, or settingShift + Cmd + O(Windows and Linux:Shift + Ctrl + O): start a new chat
For the full list, see FAQ, glossary, and shortcuts.
Common questions
What kinds of files can I upload, and is there a size limit?
You can upload documents (including scanned PDFs), spreadsheets, slides, images, audio, video, archives (ZIP files are unpacked automatically), and code or logs, with as many files as you want in a single message. Aleria can handle very large files such as full annual reports or hours of audio, and if something is too big it tells you and suggests splitting.
Does the Enhance button send my message for me?
No. Enhance only rewrites your draft in place into a clearer, more specific version. Nothing is sent until you press send, so you always review the final wording first and can accept it, edit it, or keep your own.
What is the difference between voice input and recording a meeting?
Voice input is dictation for a single message: press the microphone and Aleria transcribes as you speak, in English, Arabic, and many other languages. To capture a whole meeting instead of one message, use the Record meeting button next to the input.
How do I check the figures Aleria gives me?
When Aleria states a factual claim such as a number, date, or quote, it adds an inline citation. Click the citation marker to open a card showing the exact source (a file, a database row, a web page, or a connected system), then open that source to verify it yourself.
Which model should I pick, and what does the thinking toggle do?
Most of the time you can let Aleria pick the right model for the job, or use the model selector near the chat input to choose a faster model for quick drafts or a stronger one for complex analysis. The thinking toggle makes the model reason more carefully before it answers, which helps on trickier tasks.