MCP and AI
MCP servers
On this page
Bizisy runs two MCP servers. Every action is a tool on the server of its audience, so an AI assistant can do in Bizisy what its person can do, with their permissions, and nothing else.
| Server | Address | Tools | For |
|---|---|---|---|
| HR Assistant | /mcp/manage | 1096 | Owners, admins, HR and viewers: what their role allows in Manage. |
| My Workplace | /mcp/me | 389 | Everyone: their own details, the directory and the org chart. |
Addresses#
| Connect with | Manage | Me |
|---|---|---|
| Signing in (OAuth) | https://<company>.bizisy.com/mcp/manage | https://<company>.bizisy.com/mcp/me |
| An API key | https://api.bizisy.com/mcp/manage | https://api.bizisy.com/mcp/me |
Both use the Streamable HTTP transport. Signing in is the way for people (no key to copy, nothing to paste but the address); keys suit apps that can't sign in, and your own agents.
Fewer tools: toolsets and discovery#
/mcp/manage lists every tool the person may use: well over a hundred for HR. Long tool lists cost an assistant context on every request and make it pick worse. Add toolsets to the address to list only what a task needs:
| Address | Lists |
|---|---|
…/mcp/manage | Every tool (unchanged) |
…/mcp/manage/people | The People toolset, plus the core tools |
…/mcp/manage/people+jobs+pay | Those toolsets, plus the core tools |
…/mcp/manage/discover | Only the core tools: the assistant finds the rest when it needs them |
The toolsets: people, jobs, structure, positions, titles, pay, approvals, checklists, imports, home, documents and settings (the same on /mcp/me, with the tools a person has there). The core is platform_whoami, platform_skills_get, and two discovery tools:
platform_tools_search: finds tools of the server by words ("hire someone", "move people to a new team") or bytoolset, best match first, with each tool's input schema and whether it writes.platform_tools_call: runs any tool of the server bynamewith itsarguments(anddry_run,idempotency_key), with the same permissions, Activity and results as calling it directly.
So an assistant connected to a toolset address can still do everything its person can. Signing in works with toolset addresses too: the connection is the server's (Manage or Me) whatever the toolsets.
What the servers give an assistant#
- Tools: one per action the person's role allows, named like the action with underscores (
org.people.hire→org_people_hire). The list is filtered per caller: a viewer's assistant never seesorg_people_hire. Write tools takedry_runandidempotency_key. See Tools. - Skills: procedures written for assistants (setting up a company, managing people, webhooks, …), served as MCP resources such as
skill://org/setup-company. See Skills. - Instructions: read
platform_whoamifirst, read before writing, preview every change withdry_run, show it and commit only after the person agrees, and show the person the link to check it in Bizisy.
Results#
Tool results are JSON with the action's output and, for changes and previews, view_url: the change's entry in Activity on the company address. The text ends with "View in Bizisy: code, message and hint as the REST API, and retry_after_seconds when waiting that long and retrying may work.
The same rules as everywhere#
- Each tool call counts against the rate limit of the key or connection.
- An assistant is a machine: no other people's private or pay data, and approving, pay and a few settings stay in the web app.
- Every change, preview and failed attempt is recorded in Activity with the app's or key's name.
Something missing or wrong on this page? Write to hello@bizisy.com.