Skip to content
Tandem logo

Tandem

Runtime that checks each agent action against data scope, tool policy and approval rules before it executes, and logs the decision.

AgentsSDKSelf-hostedFor developers
tandem.ac ↗X
At a glanceswept 2026-10-12

≈ 5visits/mo

Little sign of use · estimate, read 2026-10-12

Price
Price not read
App
iPhone & iPad · 4.6★ from 38.4k ratingscurrent version 2026-10-01 ↗
Activity
Maintainedrelease 13 days ago · site updated 3 months ago
Company
founding year not on record
Runs as
SDKSelf-hosted
AlternativesSet it up for my teamIs Tandem yours? Claim the card →
What people do with it
  • gate refunds above a set amount behind human approval
  • block agents from calling tools outside their grant
  • keep an audit record of each agent action
The companyeach fact with its source · read 2026-10-12
How a program reaches itfound, not claimed · read 2026-10-13
  • MCP serverac.tandem/docs-mcpverified in the MCP registry
  • Stateanswers, 13 toolschecked 2026-10-13
  • Address to connecttandem.ac/mcp

MCP server details →

Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.

Alternatives in Agentsby use, among tools still maintained

Figure and bar: visits/mo, estimate · year: online since

All Agents tools →

Questions this page answers
  • What is Tandem?

    Runtime that checks each agent action against data scope, tool policy and approval rules before it executes, and logs the decision. It is listed in the Atlas under Agents.

  • Do people use Tandem?

    Little sign of use: ≈ 5 visits/mo · estimate, 2026-10.

  • Is Tandem still maintained?

    Maintained: release 13 days ago · site updated 3 months ago.

  • Does Tandem have an iPhone app?

    Yes: “Tandem: Practice Languages” on the App Store, rated 4.6 from 38.4k ratings; current version released 2026-10-01.

  • What are alternatives to Tandem?

    In Agents, by use: BrowserOS, Prefactor, Loop Engineering, OneCLI, Aqta, Preloop.

Not quite it? Describe the job

beta

Five tools from the Atlas that fit it, each judged against what you wrote.

How we read a tool

No votes, no reviews, no vendor claims. Every week a crawler reads each tool's own site and a few public registries, and the words on the card are bands over what it read.

  • Use: visits to the site (estimated), installs from npm and PyPI, presence in Chrome's usage report.
  • Activity: the newest release on GitHub, npm or PyPI; the newest dated page on the site; open roles on a public jobs board.
  • Price: the vendor's own pricing page, read with the date. A figure is printed only when it is on that page.
  • The line and the use cases are written by us from the homepage, one tool at a time, and refused when they repeat the vendor's marketing.

How AI assistants read each site — the reading vendors ask us about — is on each tool's own visibility page.