NewSign anything, sealed. Fill and sign any PDF in place, scans included, with every signed copy sealed.See what shipped
Developers

Build on Qasha, with the access you already have.

Connect AI agents through the MCP server, send events to your own systems with signed webhooks, and give every agent or script its own personal token.

Quick start

An agent in Qasha in three steps.

  1. Make a personal tokenIn Qasha, open Settings → API tokens & agents and create one. Copy it now: it is shown once.
  2. Add the server to your agentThe server’s address is shown, ready to copy, beside your tokens. With Claude Code, for example:
  3. Ask for what you need“Which documents are waiting for my signature?” The agent answers with waiting_for_my_signature, as you.
Terminal
claude mcp add --transport http qasha \
  https://<your-qasha-address>/api/mcp \
  --header "Authorization: Bearer $QASHA_TOKEN"

Keep the token in an environment variable, not in a file you share. Any client that can send a bearer token works; see connecting.

Personal tokens

One token, one colleague’s access.

Agents and scripts use personal tokens, sent as Authorization: Bearer <token>. There is no all-powerful key: whatever carries a token is limited to what its maker can see and do.

Made by each colleague
In Settings → API tokens & agents, named for whatever will use it.
Their access, nothing more
A token acts with its maker’s current workspace permissions, so an agent can never do more than they can.
Shown once
Copy it when it is made. Qasha keeps only a digest of it, never the token itself.
Revoked in one click
Each token shows when it was last used, and revoking it cuts off whatever carries it.
One per agent or script
Up to ten live tokens per person, so each can be revoked on its own.

Something to connect that isn’t here?

Tell us what you want Qasha to work with, and we’ll show you how it fits today.