agent-farm

A command-line tool for coordinating several coding agents while they work on the same software project. It runs builders in separate Git worktrees, which are isolated copies of a repository.

In plain words
What is it for?
Use it to start an agent workspace, launch architect and builder processes, and coordinate development across local or remote machines.
Why use it?
It helps divide development work between agents without letting their file changes interfere with one another. It also provides commands for starting and stopping the shared workspace.

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/cluesmith/codev/agent-farm
Clone the repo
git clone --depth 1 https://github.com/cluesmith/codev
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 7,100 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 $0.07100
Opus 5 $0.00000 $0.03550
Sonnet 5 $0.00000 $0.01420
Haiku 4.5 $0.00000 $0.00710

Measured yesterday against content hash 6edb58fb36d9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-farm 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 yesterday.

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.

codev-skeleton/resources/commands/agent-farm.md · 927 lines

How it starts

The opening of the file, as written. The whole thing — 927 lines — stays where its author put it; the contents beside it link to each section on GitHub.

afx - Agent Farm CLI

The afx (agent-farm) command manages multi-agent orchestration for software development. It spawns and manages builders in isolated git worktrees.

Synopsis

afx <command> [options]

Global Options

--architect-cmd <command>    Override architect command
--builder-cmd <command>      Override builder command
--shell-cmd <command>        Override shell command

Commands

afx workspace

Workspace commands - start/stop the workspace for this project.

Deprecation note: afx dash is a deprecated alias for afx workspace. It still works but prints a deprecation warning.

afx workspace start

Start the workspace.

afx workspace start [options]

Options:

  • -c, --cmd <command> - Command to run in architect terminal
  • -p, --port <port> - Port for architect terminal
  • --no-role - Skip loading architect role prompt
  • --no-browser - Skip opening browser after start
  • -r, --remote <target> - Start Agent Farm on remote machine (see below)
  • --allow-insecure-remote - Bind to 0.0.0.0 for remote access (deprecated)

Description:

Starts the workspace with:

  • Architect terminal (Claude session with architect role)
  • Web-based UI for monitoring builders
  • Terminal session management (shellper processes)

The workspace overview is accessible via browser at http://localhost:<port>.

Examples:

# Start with defaults
afx workspace start

# Start with custom port
afx workspace start -p 4300

# Start with specific command
afx workspace start -c "claude --model opus"

# Start on remote machine
afx workspace start --remote user@host
Remote Access

Start Agent Farm on a remote machine and access it from your local workstation with a single command:

# On your local machine - one command does everything:
afx workspace start --remote user@remote-host

# Or with explicit project path:
afx workspace start --remote user@remote-host:/path/to/project

# With custom port:
afx workspace start --remote user@remote-host --port 4300

Read the full file on GitHub · 927 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. yesterday First seen · 927 lines · 0 tokens per session scan A 6edb58fb36d9

Subscribe to this mod's changes

agent-farm is a command published in the GitHub repository cluesmith/codev (286 stars, last pushed 5d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 7,100 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.