Happy slice of mcp pizzaMCP.pizza

axe

MCP.Pizza Chef: jrswab

Each agent you build does one job — reviewing a pull request or summarizing a log file — and runs instantly from your command line, a script, a git hook, or a scheduled task. You describe the job in a small text file, then run the agent by name and feed it text through your terminal. It works with Claude, GPT-4o class models, Gemini, or a model running on your own computer, and can reach out to other connected tools too. Setup means installing a small program and adding a personal key from your AI provider — there's no visual app to click through.

Coding
Data
Writing

Use This MCP client To

Review my code changes before I open a pull request Summarize a long error log into a few key lines Write a commit message from my staged changes Run the same small AI task automatically on a schedule Trigger an AI check every time I commit code Chain a few small AI helpers together for one task

README

Axe

axe banner

A CLI tool for managing and running LLM-powered agents.

Why Axe?

Most AI tooling assumes you want a chatbot. A long-running session with a massive context window doing everything at once. But that's not how good software works. Good software is small, focused, and composable.

Axe treats LLM agents the same way Unix treats programs. Each agent does one thing well. You define it in a TOML file, give it a focused skill, and run it from the command line. Pipe data in, get results out. Chain agents together. Trigger them from cron, git hooks, or CI. Whatever you already use. No daemon, no GUI, no framework to buy into. Just a binary and your configs.

Overview

Axe orchestrates LLM-powered agents defined via TOML configuration files. Each agent has its own system prompt, model selection, skill files, context files, working directory, persistent memory, and the ability to delegate to sub-agents.

Axe is the executor, not the scheduler. It is designed to be composed with standard Unix tools — cron, git hooks, pipes, file watchers — rather than reinventing scheduling or workflow orchestration.

Features

  • Multi-provider support — Anthropic, OpenAI, Ollama (local models), OpenCode, and AWS Bedrock
  • TOML-based agent configuration — declarative, version-controllable agent definitions
  • Sub-agent delegation — agents can call other agents via LLM tool use, with depth limiting and parallel execution
  • Persistent memory — timestamped markdown logs that carry context across runs
  • Memory garbage collection — LLM-assisted pattern analysis and trimming
  • Skill system — reusable instruction sets that can be shared across agents
  • Stdin piping — pipe any output directly into an agent (git diff | axe run reviewer)
  • Local agent directories — auto-discovers agents from <cwd>/axe/agents/ before the global config, or use --agents-dir to point anywhere
  • Dry-run mode — inspect resolved context without calling the LLM
  • JSON output — structured output with metadata for scripting
  • Built-in tools — file operations (read, write, edit, list) sandboxed to working directory; shell command execution; URL fetching; web search
  • Output allowlist — restrict url_fetch and web_search to specific hostnames; private/reserved IPs are always blocked (SSRF protection)
  • Token budgets — cap cumulative token usage per agent run via [budget] config or --max-tokens flag
  • MCP tool support — connect to external MCP servers for additional tools via SSE or streamable-HTTP transport
  • Configurable retry — exponential, linear, or fixed backoff for transient provider errors (429, 5xx, timeouts)
  • Minimal dependencies — four direct dependencies (cobra, toml, mcp-go-sdk, x/net); all LLM calls use the standard library

Installation

Requires Go 1.25+.

axe FAQ

Can I use this to review my code before committing?
Yes — set up a reviewer agent and pipe your changes straight into it from the command line so it checks your work every time.
Can I use this to summarize long files or logs?
Yes — feed a text file or command output to a summarizing agent and it hands back the key points.
Do I need an app like Claude Desktop or Cursor to use this?
No — it runs entirely from your computer's command line rather than inside a chat app, so it suits people comfortable typing commands.
Do I need an account or key with an AI company?
Yes — you need an account and a personal key with a provider such as Anthropic, OpenAI, or Google, then add that key to your settings.
Is this easy to set up if I'm not a programmer?
Not really — it's built for people comfortable installing command-line programs, and there is no point-and-click setup screen.
Can it work with an AI model running on my own computer instead of the cloud?
Yes — it supports local models through Ollama, so nothing has to leave your machine if you set it up that way.
Can I have it run automatically without me typing a command each time?
Yes — you can trigger it from a scheduled task, a git hook, or any other automated trigger your computer already runs.
Can I connect it to other tools for extra abilities?
Yes — it can reach out to other connected tools for things like fetching web pages, alongside Claude, GPT-4o class models, or Gemini.