# Agentwork

> One place to ask, across every tool.

Agentwork connects to the tools your team already uses and answers questions across all of them in plain language with sources you can check. When the answer isn't written down, it asks the person who knows.

## What Agentwork does

- **Sourced answers** — Every response links back to the exact sources, so you can verify it in one step. When two sources contradict each other, Agentwork flags the conflict and resolves it with the right person instead of silently picking one.
- **Plain-language questions** — Anyone can ask like they'd ask a coworker and get a clear answer in seconds. No digging, no waiting on the person who knows.
- **Knowledge that improves** — When the answer isn't there yet, Agentwork spots the gap, asks whoever knows, and saves it for next time.
- **Repeatable workflows** — Schedule recurring briefings and tasks like a daily overview built for your role.
- **Shared and private knowledge** — Shared sources everyone can see, plus private ones each employee connects through their own account. Private data never surfaces in someone else's answer; the boundary is enforced in code, not the model's judgment.
- **Permission-aware access** — Agentwork respects the same access rules your tools already enforce. People only get answers drawn from sources they already have permission to see.

## How it works

1. **Integrate** — Plug Agentwork into the systems where your work lives. It connects every tool and turns scattered information into one place your company can draw from.
2. **Ask** — Anyone can ask a question in plain language and get a clear, trustworthy answer in seconds.
3. **Improves** — When the answer isn't there yet, Agentwork spots the gap, asks whoever knows, and saves it for next time.

## Integrations

Agentwork connects to Slack, Notion, Google Workspace (Drive, Docs, Sheets, Slides, Gmail, and Calendar), Microsoft 365 (Outlook mail and calendar), Fireflies, GitHub, GitLab, Linear, Atlassian (Jira and Confluence), HubSpot, Attio, Stripe, Shopify, PostHog, Mixpanel, Google Ads, LinkedIn Ads, Granola, Pleo, Numina, Oneleet, n8n, and Semrush.

You can also connect AI tools to Agentwork through the Model Context Protocol (MCP). We work on adding new integrations every day.

## Security and compliance

- Data encrypted in transit. Integration credentials are never stored in plaintext. End-to-end encryption for data at rest is in progress.
- GDPR compliant, with strict access controls on data synced from connected tools.
- Working toward SOC 2 and ISO 27001 certification (estimated completion September 2026).
- Your data is only processed to answer your team's questions, never to train shared models. No Agentwork employee can browse or read your content. Disconnect any source or delete everything at any time.
- Sub-processors are deliberately kept small and exclusively EU-based.

## FAQ

### Which tools does Agentwork connect to?

Agentwork connects to Slack, Notion, Google Workspace (Drive, Docs, Sheets, Slides, Gmail, and Calendar), Microsoft 365 (Outlook mail and calendar), Fireflies, GitHub, GitLab, Linear, Atlassian (Jira and Confluence), HubSpot, Attio, Stripe, Shopify, PostHog, Mixpanel, Google Ads, LinkedIn Ads, Granola, Pleo, Numina, Oneleet, n8n, and Semrush. You can also connect AI tools to Agentwork through the Model Context Protocol (MCP). We work on adding new integrations every day.

### What happens when the answer isn't written down anywhere?

It doesn't guess. Agentwork works out who is most likely to know and sends them one short, specific question. Their answer is saved to its memory, so the next person who asks gets it instantly and nobody has to chase it down again.

### How do I know an answer is correct?

Every answer comes with its sources, so you can click through and verify in one step. When two sources contradict each other, Agentwork flags the conflict and resolves it with the right person instead of silently picking one.

### Will people see answers they're not supposed to?

No. Agentwork respects the same access rules your tools already enforce. People only ever get answers drawn from sources they already have permission to see, so nothing leaks across teams.

### Does our company's data live on your servers?

Your data is encrypted in transit and is only ever processed to answer your team's questions, never to train shared models. Integration credentials are never stored in plaintext. We're working toward end-to-end encryption for data at rest. No Agentwork employee can browse or read your content, and you can disconnect any source or delete everything at any time.

### Who are your sub-processors?

We deliberately keep our supply chain small and exclusively EU-based. Every sub-processor we rely on operates within the EU, so your data never leaves European infrastructure. A current list is available in our DPA on request.

## Agentwork for agents

Hosted MCP: https://mcp.agentwork.com/ — ask Agentwork a question or hand it a task and get the finished, cited result back (end to end, no turn-by-turn interaction). OAuth for ChatGPT / claude.ai custom connectors, or API-key auth (`Authorization: Bearer aw_live_…`) for clients that send headers. Compatible with coding agents such as Claude, Cursor, ChatGPT, Copilot, Gemini, and Mistral.

## Links

- Homepage: https://agentwork.com/
- Sign up: https://agentwork.com/signup
- Blog: https://agentwork.com/blog (LLM-readable feed: https://agentwork.com/blog/llms.txt, RSS: https://agentwork.com/blog/rss.xml)
- Book a demo: https://savvycal.com/David-Wind-b585dcf7/agentwork
- Trust center: https://trust.agentwork.com
- Auth for agents: https://agentwork.com/auth.md
- Agent guide: https://agentwork.com/docs/agent-guide
- API catalog: https://agentwork.com/.well-known/api-catalog
- MCP server card: https://agentwork.com/.well-known/mcp/server-card.json
- Agent skills: https://agentwork.com/.well-known/agent-skills/index.json
- MCP server: https://mcp.agentwork.com/