agent-cli-dev

A development skill for starting AI coding agents in separate Git worktrees, which are isolated working copies of a repository. It supports assigning tasks to agents that can work in parallel.

In plain words
What is it for?
Use it to delegate independent features, split a large refactor, or have separate agents handle tests and implementation. It creates a new branch and isolated worktree for each agent.
Why use it?
It keeps parallel changes separate, reducing conflicts between agents and allowing independent tasks to proceed at the same time. It also explains when parallel work adds unnecessary overhead.

Skill for Claude CodeCodex

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 skills/mindroom-ai/mindroom/agent-cli-dev
Any agent
npx skills add mindroom-ai/mindroom --skill agent-cli-dev
Clone the repo
git clone --depth 1 https://github.com/mindroom-ai/mindroom

Made for: Claude Code, Codex.

Per session 49 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,092 The whole file, excluding the scripts and references it only reads on demand.
Security scan D 2 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 $0.00049 $0.01092
Opus 5 $0.00024 $0.00546
Sonnet 5 $0.00010 $0.00218
Haiku 4.5 $0.00005 $0.00109

Measured 2d ago against content hash 54fe22c81fea, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade D, and why

agent-cli-dev scanned grade D with 2 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 2d 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.

Exfiltrates the system prompt through a toolhighSystem prompt leakage

Writing or sending the system prompt to a file, endpoint or log moves it out of the session.

# 1. Write prompt to file (Claude does this with the Write tool)

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

agent-cli dev run <branch-name> cat .claude/REPORT.md
.claude/skills/agent-cli-dev/SKILL.md · 144 lines

How it starts

The opening of the file, as written. The whole thing — 144 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Parallel Development with agent-cli dev

This skill teaches you how to spawn parallel AI coding agents in isolated git worktrees using the agent-cli dev command.

Installation

If agent-cli is not available, install it first:

# Install globally
uv tool install agent-cli

# Or run directly without installing
uvx agent-cli dev new <branch-name> --agent --prompt "..."

When to spawn parallel agents

Spawn separate agents when:

  • Multiple independent features/tasks can be worked on in parallel
  • Tasks benefit from isolation (separate branches, no conflicts)
  • Large refactoring that can be split by module/component
  • Test-driven development (one agent for tests, one for implementation)

Do NOT spawn when:

  • Tasks are small and sequential
  • Tasks have tight dependencies requiring constant coordination
  • The overhead of context switching exceeds the benefit

Core command

For short prompts:

agent-cli dev new <branch-name> --agent --prompt "Fix the login bug"

For longer prompts (recommended for multi-line or complex instructions):

agent-cli dev new <branch-name> --agent --prompt-file path/to/prompt.md

This creates:

  1. A new git worktree with its own branch
  2. Runs project setup (installs dependencies)
  3. Opens a new terminal tab with an AI coding agent
  4. Passes your prompt to the agent

Important: Use --prompt-file for prompts longer than a single line. The --prompt option passes text through the shell, which can cause issues with special characters (exclamation marks, dollar signs, backticks, quotes) in ZSH and other shells. Using --prompt-file avoids all shell quoting issues.

Writing effective prompts for spawned agents

Spawned agents work in isolation, so prompts must be self-contained. Include:

  1. Clear task description: What to implement/fix/refactor
  2. Relevant context: File locations, patterns to follow, constraints
  3. Report request: Ask the agent to write conclusions to .claude/REPORT.md

Read the full file on GitHub · 144 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 2d ago First seen · 144 lines · 49 tokens per session scan D 54fe22c81fea

Subscribe to this mod's changes

agent-cli-dev is a skill published in the GitHub repository mindroom-ai/mindroom (254 stars, last pushed 2d ago), licensed Apache-2.0. It adds 49 tokens to every session and 1,092 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it D with 2 findings (exfiltrates the system prompt through a tool, reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens