# Actions with Receipts

> How the board goes from advice to execution — proposed actions, Skeptic risk review, explicit approval, a 30-second undo window, and a receipt for everything that ran.

**Canonical URL:** https://www.synthboard.ai/docs/actions
**Markdown source:** https://www.synthboard.ai/docs/actions.md

## The idea

Advice that stops at advice leaves the follow-through to you. With integrations connected (62 providers across 22 categories — Gmail, Slack, Linear, Notion, HubSpot and more), your board proposes the follow-up — draft the reply, open the ticket, update the record — and every action runs through the same governance pipeline: reviewed, approved by you, undoable, and receipted.

## The governance pipeline

Four gates stand between a synth's idea and anything actually happening:

1. **Propose.** A synth proposes the action inside its response — grounded in the context it just argued from (the Gmail thread it read, the Linear ticket it cited).
2. **Skeptic review.** Before the proposal ever reaches you, the Skeptic reviews it for risk. Destructive actions always go through this review.
3. **Your approval.** You see a clean preview of exactly what will run — recipient, content, target record — and nothing executes until you approve it.
4. **Execute, with undo.** The action runs, and you get a 30-second undo window to reverse it on the spot.

## The receipt

Each action that runs is logged with its full provenance: which synth proposed it, what the Skeptic flagged, when you approved it, what actually executed, and whether it was undone. The session record keeps these receipts alongside the memo — the decision *and* its follow-through live in one place.

When an AI can touch your email, your CRM, and your issue tracker, "trust us" is not a governance model. A receipt for every action — plus an undo — is.

## Your controls

- **Disable per session** — run any session with tool execution off entirely.
- **Restrict by tool** — allow only specific tool slugs (for example: Gmail drafts only, Linear reads only). Over the API and MCP the same control ships as `integrations_enabled` and `allowed_tool_slugs` on the session config.
- **Reads vs. writes** — reading context (a thread, a ticket, a deal) and proposing an action are separate steps; connecting an integration never grants silent write access.

## Related

- Integrations: https://www.synthboard.ai/docs/integrations
- Security posture: https://www.synthboard.ai/docs/security
- Your World Model: https://www.synthboard.ai/docs/world-model