Atlas / MCP servers / toloka / Tendem

TendemSAFE

mcp/toloka/tendem

Home for Human In The Loop Tendem plugin for your agent.

Verdict
SAFE
Grade
B
Trust score
89 /100
Exposed tools
—
Transport
streamable-http
License
MIT
Stars
102
01

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

Delegate tasks to vetted human experts — straight from your AI assistant. 🧑🔬🤝🤖

Tendem gives your AI direct access to vetted human experts for the work an agent alone can't nail: judgment-heavy research, competitive analysis, copywriting and editing, design review, data cleaning, and complex multi-step tasks. You (or your agent) submit a task in plain English — Tendem scopes the work, quotes a transparent price, and delivers verified results back into the chat as files, reports, or structured outputs. Agent plus expert raises the quality ceiling of what you can ship well beyond what either does alone.

Ideal for founders, creators, operators, and consultants who need reliable expert output without managing freelancers.

Get started → [agent.tendem.ai](https://agent.tendem.ai).

This repo is the official distribution home for the Tendem MCP plugin — the part that lives in your AI client. Install it and your assistant arrives already knowing how to drive the full Tendem lifecycle: create → scope → approve → fetch, with clean polling and spend safety built in.

The MCP server itself is hosted at https://mcp.tendem.ai/mcp — it is not in this repo. This repo contains the client-side plugins, skills, and rules that teach your AI client to use the hosted server well.

Start using Tendem

1. The web interface

Sign in at [agent.tendem.ai](https://agent.tendem.ai) to create your account, top up your balance, and watch tasks run end-to-end. The full output of every task — markdown plus downloadable files — is always available there, no matter which client you launched it from.

2. Install the plugin in your AI client

The hosted server URL is the same everywhere: https://mcp.tendem.ai/mcp (streamable HTTP, OAuth on first use). Pick your client:

Claude Code

/plugin marketplace add Toloka/tendem-mcp
/plugin install tendem@tendem-mcp

##

Read from source at commit a60f4fa6a718OBSERVED · 2026-10-07
02

Connect

Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.

claude-code
claude mcp add langchain-tendem --env OPENAI_API_KEY=${OPENAI_API_KEY} --env TENDEM_API_KEY=${TENDEM_API_KEY} -- uvx langchain-tendem
claude-desktop
{
  "mcpServers": {
    "langchain-tendem": {
      "command": "uvx",
      "args": [
        "langchain-tendem"
      ],
      "env": {
        "OPENAI_API_KEY": "${OPENAI_API_KEY}",
        "TENDEM_API_KEY": "${TENDEM_API_KEY}"
      }
    }
  }
}
03

Trust audit

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

LayerWhat it checksResult
L0Provenance & inventoryPASS
L1Static analysis of the codePASS
L2Instruction surface (what it tells the agent)PASS
L3Class-specific surfacePASS
L4Behavioural (sandbox)SKIPPED

What the source does

Filesystem
none-observed
Network
declared (6 observation(s))
Shell
none-observed
Dependencies
pinned
Secrets in source
none-found

Findings (2)

LOWInventory / provenance · inv.hidden_file · CWE-1104
.gitmodules
.gitmodules
Why it matters. hidden member outside the usual dotfiles
Fix. review its purpose
LOWInventory / provenance · mcp.no_tools_extracted · CWE-1104
server.json
Why it matters. the tool list is enumerated at runtime by tools/list, not declared in source
Fix. the page says so rather than showing an empty table

Gates applied: no_behavioural_pass.

Audited 2026-10-07 · audit v0.4.1 · source sha a60f4fa6a718full audit observations/trust-audit/mcp-server/toloka__tendem.json · Report an issue / request a re-scan
04

Audit history

Every audit this server has had. A grade with a past is a grade somebody is still checking.

DateSourceVerdictGradeScoreChange
2026-10-07a60f4fa6a718SAFEB89first audit
05

Questions

What is the Tendem MCP server?

Home for Human In The Loop Tendem plugin for your agent.

Is Tendem safe to connect to an agent?

The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.

What credentials does Tendem need?

It reads OPENAI_API_KEY and TENDEM_API_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.

How does Tendem run?

It speaks streamable-http, so it runs as a service you connect to over the network. It is published on PyPI as langchain-tendem.

How current is this page?

The grade is for one exact copy of the source (a60f4fa6a718), read on 2026-10-07. The repository is watched and re-audited when it changes.

Advertisement