Raq.com brings 70+ business and AI tools into a single account. Your team works in the browser. Codex, Claude Code and Raq’s own Supercomputer connect through MCP. Everything they do stays on the record.
Free trial · no card required · every tool included in every plan
Why Raq exists
A chat product here, a task board there. A signing service, a form builder, a file-sharing account, a monitoring service and several AI subscriptions — each with its own login, billing, permissions and audit history.
An AI agent becomes far less useful when it can see only one piece of that setup.
Raq.com puts those jobs in one account and gives agents a common route into them. A task can start with a message, use a file from File Drop, create a document, request approval, update a board — and leave the result where the whole team can inspect it.
How it works
The browser and the agent interface work against the same account, the same files and the same tasks. A person can amend work created by an agent; an agent can continue work started by a person.
Your team writes messages, manages boards, uploads files, reviews generated media, approves requests and edits documents through ordinary interfaces.
Raq exposes structured operations through the Model Context Protocol — list boards, create a card, upload a file, run an image batch, read a project record.
Describe a job in plain English and Raq’s own agent chooses and runs the right tools — no coding environment to configure. It works with the same account, files and team as any connected external agent.
Prepare a briefing on our top three competitors and share it with the sales channel.
Market Intel researching named companies… Pretty Docs formatting the briefing… Team Comms posting to #sales… ✓ Done — briefing filed to Project Base for review.The catalogue
Ordinary business software and specialist AI production, side by side — so an AI result can enter a normal company process instead of ending as text in a chat window.
Channels, team messaging, voice calls, personas and AI-assisted search.
Kanban boards with dependencies, time tracking and public intake forms.
Project hubs with versioned notes, structured tables and linked boards.
A daily view separating human, mixed and agent-owned work.
Team articles with imports, AI search and summaries.
A shared Markdown notepad with live editing and version history.
File storage and sharing, with search and links into projects and boards.
Staff measures, daily evidence and material for performance reviews.
One place to compare image, video and text options submitted by agents.
Daily entries with highlighted days and replies by email or Telegram.
Leads, contacts and pipeline stages in a shared account.
Branded, magic-link workspaces where clients view, upload and approve.
Shared inboxes, email-linked tickets and AI-assisted replies.
Support assistants for your site, with testing and visitor insights.
Branded forms with 17 question types and response reporting.
A brief becomes a plan, forms, copy, automations and results.
Platform-specific posts published through connected accounts.
Monitors companies or sectors and prepares research briefings.
Approved AI assistants make business calls, recorded and transcribed.
Freeze a web page for pinned written or voice feedback.
Electronic signatures with templates, bulk sends and an audit record.
Turns text, DOCX or PDF input into designed documents.
Multi-model contract review with clause comments and recorded risk.
Writes sentence by sentence with ranked alternatives from leading models.
Several rewrites or drafts from one description, down to sentence level.
Passwords and API keys shared through encrypted links that expire.
Credentials released to agents one at a time after biometric approval.
Reads batches of receipts into dates, merchants, taxes and line items.
Documents, images, audio and video conversion, including OCR export.
Compares text versions with named snapshots and inline comments.
One conversation, several models — your chosen answer continues the thread.
One question to leading models, combined into a single answer.
Longer, structured reasoning reports.
Several models run a blind deliberation and produce a verdict.
Define prompts, run models, rate answers, publish a leaderboard.
Reusable model personalities, tones and system instructions.
Versioned prompts with variables, reusable blocks and tags.
Versioned instructions that guide Supercomputer across sessions.
Image batches across models, with prompt variants and source guidance.
Video from text, images or references, across models and price points.
Music and sound effects with prompt variants and multiple outputs.
Animated SVG work, exported as SVG, WebP or GIF.
Many HTML design alternatives from several models — pick the strongest.
Backgrounds, shadows and device frames, exported at chosen sizes.
Record screen or webcam in the browser and share with timed comments.
Extracts a brand from your website into a shared guidelines workspace.
Audio and video to text with word timestamps and SRT export.
Speech from text, or a changed voice in an existing recording.
AI-agent website design and hosting through 102.ai.
Availability checks from several regions, with incident records.
Watches pages, feeds or sitemaps and records exact changes.
Server heartbeat and load data through lightweight agents.
Search performance, answer-engine readiness and AI visibility.
Companies House financials, officers, filings and ownership data.
Creates or accepts spreadsheets, then transforms and enriches the data.
Tracked links on raq.link and raq.to.
Branded QR codes with chosen colours and logos.
Raq’s built-in agent — describe the job in plain English.
Multi-step processes across tools, on events or schedules.
Repeatable workflows across managed Codex and Claude sessions.
Supervise terminal sessions from a browser, away from the desk.
A multi-terminal control room with keyboard controls and a live queue.
Structured requests between people and agents, with two-stage approvals.
Your coding agent builds sandboxed applications inside Raq.
Request the tool you’re missing, or register interest.
Continuity
Because the tools share an account, a job doesn’t stop at each product boundary. Here’s client onboarding, end to end — people and agents taking turns on the same record.
A prospect completes your branded intake form.
Updates the CRM record and opens an onboarding card.
A branded, magic-link workspace is created for the client.
The agreement goes out for signature with a full audit record.
Signed documents and client uploads land in the account’s files.
The board and channels keep the discussion and next actions.
Brand Guidelines holds the rules. Marketing Campaigns plans, Ghostwriter and the Image and Video Labs produce, Creative Review collects decisions, Social Scheduler publishes after approval.
Website Build & Host ships the site. SEO & AEO Checker reviews it, Site Uptime and Page Watch monitor it, and Help Desk plus Website Chatbox AI handle incoming support.
Record the outcome in Priorities, let Orchestrator run the workflow, supervise through Nature Work or Agent Operator HQ, and approve boundary decisions with Ask My Agent.
Agent operations
Longer agent jobs need execution, supervision, approvals, hand-offs — and a record left behind. Raq treats agent operations as a first-class discipline.
Runs templated, repeatable workflows across managed Codex and Claude sessions.
Connects terminal sessions to a browser-based viewer, so you can supervise work away from the desk.
Several agent terminals in one command view, with keyboard controls and a live message queue.
Separates human, mixed and agent-owned tasks so you can see exactly where attention is needed.
orchestrator run onboarding-pack
workflow · 2 sessions · account: harbour-and-co
claude-1 building client portal pages… done 2m14s
codex-2 drafting campaign copy → Creative Review done 1m48s
claude-1 filing outputs to Project Base… done
⏸ ask-my-agent approval required — send services agreement?
waiting for operator · two-stage approval · audit logged
Extend the platform
Build it yourself with your own coding agent, or have Raq design, build, host and maintain it for you. Both routes sit beside the shared tools — and responsibility for the build stays clear.
Point your coding agent at Raq and let it build an application that runs inside your account — sandboxed, permissioned and shareable with your team.
Describe the workflow — that awkward spreadsheet, that internal process — and Raq designs, builds, hosts and maintains a native module for your account.
Pricing
Per-user pricing, with every public tool in every paid plan. Higher plans bring a larger monthly AI allowance and higher service limits.
£14per user / month, excl. VAT
For small teams replacing their first stack of separate subscriptions.
£45per user / month, excl. VAT
For teams putting AI and agents to work every day.
£200per user / month, excl. VAT
For companies running AI production and agent workflows at volume.
£500per user / month, excl. VAT
For accounts running heavy, continuous agent operations.
Privacy & security
Core text and chat AI uses Zero Data Retention routes by default — the model provider cannot keep prompts or responses for training. Raq.com does not train models on customer content.
Image, video and voice tools use specialist providers under their own retention terms, with completed media downloaded to Raq’s own storage. Settings for AI privacy and live web search stay in your hands.
Assessing Raq for regulated or sensitive data? Read the current Privacy Policy, AI Privacy & Security page and Subprocessors list.
Get started
Try every tool free, no card required. Connect Claude Code or Codex in minutes, or just ask Supercomputer.
From £14 per user per month · every tool in every plan