agentleFS
Sign inStart free
directionally-localdirectionally-local/autopilot/docs/llms.txt

autopilot

llms.txt177 starsChanged 3 days agoMITView on GitHub

kimiflare is an open-source, terminal-based AI coding agent — a self-hosted alternative to Claude Code, Aider, Gemini CLI, and Codex CLI — that runs on your own Cloudflare account. It is powered by the Kimi K2.7 model on Cloudflare Workers AI, with every request routed through your own Cloudflare AI Gateway for per-request logs, response caching, and authoritative cost. It ships as the kimiflare npm package (npm install -g kimiflare), needs no API keys beyond your Cloudflare credentials, and is…

  • Installs packages
Create a free account to mount
Files
llms.txt
# kimiflare

> kimiflare is an open-source, terminal-based AI coding agent — a self-hosted alternative to Claude Code, Aider, Gemini CLI, and Codex CLI — that runs on your own Cloudflare account. It is powered by the Kimi K2.7 model on Cloudflare Workers AI, with every request routed through your own Cloudflare AI Gateway for per-request logs, response caching, and authoritative cost. It ships as the `kimiflare` npm package (`npm install -g kimiflare`), needs no API keys beyond your Cloudflare credentials, and is MIT-licensed.

## What kimiflare is

- **Category:** AI coding agent / agentic coding harness / terminal coding assistant (CLI).
- **Model:** Kimi K2.7 (`@cf/moonshotai/kimi-k2.7-code`) on Cloudflare Workers AI, 262K-token context window.
- **Runs on:** your own Cloudflare account (bring-your-own-cloud / BYOK). All inference and logs stay in your account.
- **Closest comparisons:** Anthropic Claude Code, Aider, Google Gemini CLI, OpenAI Codex CLI, Cursor. kimiflare's differentiator is being Cloudflare-native, open source, and self-hosted with authoritative cost from AI Gateway logs.
- **Install:** `npm install -g kimiflare` then run `kimiflare`, or `npx kimiflare`. Requires Node.js >= 20.
- **License:** MIT. **Author:** Sina Meraji (@sinasanm).

## Key features

- Plan / Edit / Auto permission modes (read-only research, per-action approval, full autonomy).
- AI Gateway observability by default: per-request logs, response caching, per-turn cost and latency, per-feature `/cost` breakdown.
- Image understanding (PNG/JPG/WebP/GIF/BMP) inline in prompts.
- Web search, GitHub read, and headless browser tools.
- LSP semantic code intelligence (hover, go-to-definition, references, diagnostics).
- MCP (Model Context Protocol) extensibility for external tool servers.
- Explicit, opt-in cross-session memory (SQLite + embeddings).
- Session persistence with `/resume`, live multi-step task panel, streaming reasoning.

## Docs

- [Home / overview](https://kimiflare.com/): what kimiflare is, features, install, architecture.
- [Claude Code alternative](https://kimiflare.com/claude-code-alternative): why and how kimiflare compares to Claude Code.
- [FAQ](https://kimiflare.com/faq): common questions about setup, cost, models, and privacy.
- [GitHub repository](https://github.com/directionally-local/autopilot): source code, issues, releases.
- [npm package](https://www.npmjs.com/package/kimiflare): installation and versions.

## Comparisons

- [kimiflare vs Claude Code](https://kimiflare.com/vs/kimiflare-vs-claude-code)
- [kimiflare vs Aider](https://kimiflare.com/vs/kimiflare-vs-aider)
- [kimiflare vs Gemini CLI](https://kimiflare.com/vs/kimiflare-vs-gemini-cli)
- [kimiflare vs Codex CLI](https://kimiflare.com/vs/kimiflare-vs-codex-cli)

## Blog

- [Run an AI coding agent on your own Cloudflare account](https://kimiflare.com/blog/run-ai-coding-agent-on-your-own-cloudflare-account)
- [What is an AI coding harness?](https://kimiflare.com/blog/what-is-an-ai-coding-harness)

Download now (ZIP)

  • Other versions1 project uses a version of this.

    Word for word · 1

    1. sinameraji/autopilot★ 177100% same

    Edited · 0

    None found.

    Wrote their own version · 0

    None found.

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repositoryOne other file it gives its agents.

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free