team

A command for running a small group of AI agents in one process, connected to shared workspace files, human approvals, shared memory, and member dispatch. The setup can come from a team package or a built-in document-review example.

In plain words
What is it for?
Running document-review flows, drafting and publishing files, requesting reviewer decisions, recording verdicts, and reading shared team memory.
Why use it?
It provides a local environment for testing how several agents collaborate on a review task and share work.

Command

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/paiml/rust-mcp-sdk/team
Clone the repo
git clone --depth 1 https://github.com/paiml/rust-mcp-sdk
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 867 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.00867
Opus 5 $0.00000 $0.00434
Sonnet 5 $0.00000 $0.00173
Haiku 4.5 $0.00000 $0.00087

Measured 3d ago against content hash 67b4507b54a9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

team scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

cargo-pmcp/docs/commands/team.md · 88 lines

How it starts

The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.

cargo pmcp team

Run an in-process small team of agents plus the reference team servers, locally.

Usage

cargo pmcp team <SUBCOMMAND>

Description

team dev composes a small team in one process — member agents wired to the four pmcp-team-servers reference servers with dev-grade backends:

Server Role
team-fs a shared workspace filesystem (draft → review)
approval-mcp human sign-off (ask / resolve)
mem-mcp shared team memory
team-mcp agent-facing member dispatch

The team is wired from a TeamPackage--package (+ --data-dir) or the built-in two-member doc-review fixture (the locked default). Composition is delegated entirely to the pmcp-team-servers runtime; team dev is a thin CLI over it, not a re-implementation.

Subcommands

Subcommand Description
dev Run an in-process small team (member agents + the four reference servers)

team dev

Three behaviors:

  • default — compose the team over in-memory transports with an offline source and print a labeled transcript of the 7-step doc-review flow (draft → publish → ask reviewer → record verdict → read → remember → dispatch), then tear down cleanly. Fully offline and deterministic — no network, no LLM, no sockets.
  • --serve — expose team-mcp over HTTP on 127.0.0.1:<port> (via the shipped team-mcp serve recipe), running until Ctrl-C. Point any MCP client at it.
  • --llm <endpoint> — swap the offline source for an OpenAI-compatible source so members are backed by a real model.
cargo pmcp team dev [--serve] [--llm <URL>] [OPTIONS]

Options

Option Description
--package <PATH> Team package to run (defaults to the built-in doc-review fixture)
--data-dir <DIR> Directory for team-fs / mem-mcp state and member AgentPackages
--serve Serve team-mcp over HTTP instead of running the in-process transcript
--port <PORT> Port for the HTTP serve path (default 8080)
--llm <URL> LLM endpoint for members backed by a real model (swaps the offline source)
--model <MODEL> Model id passed to the LLM source (--llm only; default llama3.2)
--llm-api-key-env <VAR> Environment variable holding the LLM API key (--llm only)
--allow-insecure-http Allow a plain-HTTP (non-TLS) LLM endpoint (--llm only)

Read the full file on GitHub · 88 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 88 lines · 0 tokens per session scan A 67b4507b54a9

Subscribe to this mod's changes

team is a command published in the GitHub repository paiml/rust-mcp-sdk (53 stars, last pushed 3d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 867 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.