pilot

作者 qf-studio已验证

AI that ships your tickets.

646
Stars
51
Forks
Go
语言
2026/8/23
添加时间

⚠️ 第三方软件声明

本 Skill 为第三方开源软件,独立托管于 GitHub。SkillTip 仅为信息目录,不控制或维护底层仓库。所显示的安全检查为自动化且范围有限,安装前请自行审查源码。

阅读服务条款

安装

添加到你的 Claude Code skills 目录:

# Add to your Claude Code skills
git clone https://github.com/qf-studio/pilot

快速入门

使用 pilot 等 Skills 的指南。

安全报告

已验证

上次扫描:—

{
  "status": "PASSED",
  "issues": []
}

README.md

   ██████╗ ██╗██╗      ██████╗ ████████╗
   ██╔══██╗██║██║     ██╔═══██╗╚══██╔══╝
   ██████╔╝██║██║     ██║   ██║   ██║
   ██╔═══╝ ██║██║     ██║   ██║   ██║
   ██║     ██║███████╗╚██████╔╝   ██║
   ╚═╝     ╚═╝╚══════╝ ╚═════╝    ╚═╝
  

AI that ships your tickets while you sleep

Release License: BSL 1.1 CI Go Report Card Discord

DocsInstallDesktopQuick StartHow It WorksFeaturesCLIDiscordDeploy


Pilot dashboard — live queue, autopilot PR lifecycle, execution history, git graph

The Problem

You have 47 tickets in your backlog. You agonize over which to prioritize. Half are "quick fixes" that somehow take 2 hours each. Your PM asks for status updates. Sound familiar?

The Solution

Pilot picks up tickets from GitHub, Linear, Jira, or Asana—plans the implementation, writes the code, runs tests, and opens a PR. You review and merge. That's it.

┌─────────────┐      ┌─────────────┐      ┌─────────────┐      ┌─────────────┐
│   Ticket    │ ───▶ │   Pilot     │ ───▶ │   Review    │ ───▶ │   Ship      │
│  (GitHub)   │      │  (AI dev)   │      │   (You)     │      │  (Merge)    │
└─────────────┘      └─────────────┘      └─────────────┘      └─────────────┘

Install

Homebrew (recommended)

brew tap qf-studio/pilot
brew install pilot

Go Install

go install github.com/qf-studio/pilot/cmd/pilot@latest

From Source

git clone https://github.com/qf-studio/pilot
cd pilot
make build
sudo make install-global

Desktop App

Download the desktop app from the latest release:

PlatformDownload
macOS (Universal)Pilot-Desktop-macOS-universal.dmg
Windows (x64)Pilot-Desktop-Windows-amd64-setup.exe
Linux (x64)Pilot-Desktop-Linux-amd64.tar.gz

Requirements

  • Go 1.22+ (build only)
  • Claude Code CLI 2.1.17+
  • OpenAI API key (optional, for voice transcription)

Quick Start

# 1. Initialize config
pilot init

# 2. Start Pilot
pilot start --github              # GitHub issue polling
pilot start --telegram            # Telegram bot
pilot start --telegram --github   # Both

# 3. Create a GitHub issue with 'pilot' label, or message your Telegram bot

That's it. Go grab coffee. ☕

How It Works

You label issue "pilot"
        │
        ▼
┌───────────────────┐
│  Pilot claims it  │  ← Adds "pilot/in-progress" label
└───────┬───────────┘
        │
        ▼
┌───────────────────┐
│  Creates branch   │  ← pilot/GH-{number}
└───────┬───────────┘
        │
        ▼
┌───────────────────┐
│  Plans approach   │  ← Analyzes codebase, designs solution
└───────┬───────────┘
        │
        ▼
┌───────────────────┐
│  Implements       │  ← Writes code with Claude Code
└───────┬───────────┘
        │
        ▼
┌───────────────────┐
│  Quality gates    │  ← Test, lint, build validation
└───────┬───────────┘
        │
        ▼
┌───────────────────┐
│  Opens PR         │  ← Links to issue, adds "pilot/done"
└───────┬───────────┘
        │
        ▼
    You review
        │
        ▼
      Merge 🚀

Features

133 features implemented across execution, intelligence, integrations, and infrastructure.

Core Execution

FeatureDescription
AutopilotCI monitoring, auto-merge, feedback loop (dev/stage/prod modes)
Epic DecompositionComplex tasks auto-split into sequential subtasks via Haiku API
Self-ReviewAuto code review before PR push catches issues early
Sequential ExecutionWait for PR merge before next issue (prevents conflicts)
Quality GatesTest/lint/build validation with auto-retry
Execution ReplayRecord, playback, analyze, export (HTML/JSON/MD)

Intelligence

FeatureDescription
Model RoutingHaiku (trivial) → Opus 4.6 (standard/complex), auto-detected
Effort RoutingMaps task complexity to Claude thinking depth
Research SubagentsHaiku-powered parallel codebase exploration
Navigator IntegrationAuto-detected .agent/, skipped for trivial tasks
Cross-Project MemoryShared patterns and context across repositories

Integrations

FeatureDescription
Telegram BotChat, research, planning, tasks + voice & images
GitHub PollingAuto-pick issues with pilot label
GitLab / Azure DevOpsFull polling + webhook adapters
Linear/Jira/AsanaWebhooks and task sync
Daily BriefsScheduled reports via Slack/Email/Telegram
AlertingTask failures, cost thresholds, stuck detection

Infrastructure

FeatureDescription
Dashboard TUISparkline metrics cards, queue depth, autopilot status
Persistent MetricsToken/cost/task counts survive restarts via SQLite
Hot UpgradeSelf-update with pilot upgrade or u key in dashboard
Cost ControlsBudget limits with hard enforcement
Multiple BackendsClaude Code + OpenCode support
BYOKBring your own Anthropic key, Bedrock, or Vertex

Autopilot Modes

Control how much autonomy Pilot has:

# Fast iteration - skip CI, auto-merge
pilot start --env=dev --github

# Balanced - wait for CI, then auto-merge
pilot start --env=stage --github

# Safe - wait for CI + human approval
pilot start --env=prod --github

Telegram Integration

Talk to Pilot naturally - it understands different interaction modes:

ModeExampleWhat Happens
💬 Chat"What do you think about using Redis?"Conversational response, no code changes
🔍 Question"What files handle authentication?"Quick read-only answer
🔬 Research"Research how the caching layer works"Deep analysis sent to chat
📐 Planning"Plan how to add rate limiting"Shows plan with Execute/Cancel buttons
🚀 Task"Add rate limiting to /api/users"Confirms, then creates PR
You: "Plan how to add user authentication"
Pilot: 📐 Drafting plan...
Pilot: 📋 Implementation Plan
       1. Create auth middleware...
       2. Add JWT token validation...
       [Execute] [Cancel]

You: [clicks Execute]
Pilot: 🚀 Executing...
Pilot: ✅ PR #142 ready: https://github.com/...

Send voice messages, images, or text. Pilot understands context.

Dashboard

Real-time visibility into what Pilot is doing:

┌─ Pilot Dashboard ─────────────────────────────────────────┐
│                                                           │
│  Status: ● Running    Autopilot: stage    Queue: 3        │
│                                                           │
│  Current Task                                             │
│  ├─ GH-156: Add user authentication                       │
│  ├─ Phase: Implementing (65%)                             │
│  └─ Duration: 2m 34s                                      │
│                                                           │
│  Token Usage          Cost                                │
│  ├─ Input:  124k      Today:    $4.82                     │
│  ├─ Output:  31k      This Week: $28.40                   │
│  └─ Total:  155k      Budget:    $100.00                  │
│                                                           │
│  Recent Tasks                                             │
│  ├─ ✅ GH-155  Fix login redirect      1m 12s   $0.45     │ 
│  ├─ ✅ GH-154  Add dark mode toggle    3m 45s   $1.20     │
│  └─ ✅ GH-153  Update dependencies     0m 34s   $0.15     │
│                                                           │
└───────────────────────────────────────────────────────────┘
pilot start --dashboard --github

Environment Variables

Pilot uses Claude Code for AI execution:

VariableDescription
ANTHROPIC_API_KEYCustom Anthropic API key (uses your own account)
ANTHROPIC_BASE_URLCustom API endpoint (proxies, enterprise)
CLAUDE_CODE_USE_BEDROCKSet to 1 for AWS Bedrock
CLAUDE_CODE_USE_VERTEXSet to 1 for Google Vertex AI

Example: Using AWS Bedrock

export CLAUDE_CODE_USE_BEDROCK=1
export AWS_REGION=us-east-1
pilot start --github

Configuration

Config location: ~/.pilot/config.yaml

version: "1.0"

gateway:
  host: "127.0.0.1"
  port: 9090

adapters:
  telegram:
    enabled: true
    bot_token: "${TELEGRAM_BOT_TOKEN}"
    chat_id: "${TELEGRAM_CHAT_ID}"

  github:
    enabled: true
    token: "${GITHUB_TOKEN}"
    repo: "owner/repo"
    pilot_label: "pilot"
    polling:
      enabled: true
      interval: 30s

orchestrator:
  execution:
    mode: sequential           # "sequential" or "parallel"
    wait_for_merge: true       # Wait for PR merge before next task
    poll_interval: 30s
    pr_timeout: 1h

projects:
  - name: "my-project"
    path: "~/Projects/my-project"
    navigator: true
    default_branch: main

daily_brief:
  enabled: true
  schedule: "0 8 * * *"
  timezone: "Europe/Berlin"

alerts:
  enabled: true
  channels:
    - name: telegram-alerts
      type: telegram
      severities: [critical, error, warning]

executor:
  backend: claude-code          # "claude-code" or "opencode"

CLI Reference

Core Commands

pilot start          # Start with configured inputs
pilot stop           # Stop daemon
pilot status         # Show running tasks
pilot init           # Initialize configuration
pilot version        # Show version info

pilot start

pilot start                          # Config-driven
pilot start --telegram               # Enable Telegram polling
pilot start --github                 # Enable GitHub issue polling
pilot start --linear                 # Enable Linear webhooks
pilot start --telegram --github      # Enable both
pilot start --dashboard              # With TUI dashboard
pilot start --no-gateway             # Polling only (no HTTP server)
pilot start --sequential             # Sequential execution mode
pilot start --env=stage              # Autopilot mode (dev/stage/prod)
pilot start -p ~/Projects/myapp      # Specify project
pilot start --replace                # Kill existing instance first

pilot task

pilot task "Add user authentication"                    # Run in cwd
pilot task "Fix login bug" -p ~/Projects/myapp          # Specify project
pilot task "Refactor API" --verbose                     # Stream output
pilot task "Update docs" --dry-run                      # Preview only
pilot task "Implement feature" --backend opencode       # Use OpenCode

pilot upgrade

pilot upgrade                    # Check and upgrade
pilot upgrade check              # Only check for updates
pilot upgrade rollback           # Restore previous version
pilot upgrade --force            # Skip task completion wait
pilot upgrade --no-restart       # Don't restart after upgrade
pilot upgrade --yes              # Skip confirmation

Analytics Commands

pilot brief                       # Show scheduler status
pilot brief --now                 # Generate and send immediately
pilot brief --weekly              # Generate weekly summary

pilot metrics summary             # Last 7 days overview
pilot metrics summary --days 30   # Last 30 days
pilot metrics daily               # Daily breakdown
pilot metrics projects            # Per-project stats

pilot usage summary               # Billable usage summary
pilot usage daily                 # Daily breakdown
pilot usage export --format json  # Export for billing

pilot patterns list               # List learned patterns
pilot patterns search "auth"      # Search by keyword

Architecture

┌─────────────────────────────────────────────────────────────┐
│                          PILOT                              │
├──────────────┬──────────────────────────────────────────────┤
│ Gateway      │ HTTP/WebSocket server, routing               │
│ Adapters     │ Telegram, Slack, GitHub, Jira, Linear, Asana │
│ Executor     │ Claude Code process management               │
│ Orchestrator │ Task planning, phase management              │
│ Memory       │ SQLite + cross-project knowledge graph       │
│ Briefs       │ Scheduled reports, multi-channel delivery    │
│ Alerts       │ Failure detection, cost monitoring           │
│ Metrics      │ Token usage, execution analytics             │
└──────────────┴──────────────────────────────────────────────┘

Development

make deps        # Install dependencies
make build       # Build binary
make test        # Run tests
make lint        # Run linter
make dev         # Development mode with hot reload

FAQ

Is this safe?

Pilot runs in your environment with your permissions. It can only access repos you configure. All changes go through PR review (unless you enable auto-merge). You stay in control.

How much does it cost?

Pilot is free. You pay for Claude API usage (~$0.50-2.00 per typical task). Set budget limits to control costs.

What tasks can it handle?

Best for: bug fixes, small features, refactoring, tests, docs, dependency updates.

Not ideal for: large architectural changes, security-critical code, tasks requiring human judgment.

Does it learn my codebase?

Yes. Pilot uses Navigator to understand your patterns, conventions, and architecture. Cross-project memory shares learnings across repositories.

License

Business Source License 1.1 © Aleksei Petrov

Use CaseAllowed
Internal use
Self-hosting
Modification & forking
Non-competing products
Competing SaaS❌ (requires license)

Converts to Apache 2.0 after 4 years.

Contributing

Contributions welcome. Please open an issue first for major changes.

git checkout -b feature/my-feature
make test
# Submit PR

Stop agonizing over tickets. Let Pilot ship them.

⭐ Star on GitHub

Built with Claude Code + Navigator

常见问题

What is pilot?

pilot is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by qf-studio. AI that ships your tickets. It has 646 GitHub stars.

Is pilot safe to use?

Yes. pilot passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.

How do I install pilot?

Clone the repository with "git clone https://github.com/qf-studio/pilot" and add it to your Claude Code skills directory (see the Installation section above).

What programming language is pilot written in?

pilot is primarily written in Go. It is open-source under qf-studio on GitHub, so you can review or fork the full source.

Are there alternatives to pilot?

Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh pilot against similar tools.

评论 (0)

暂无评论,成为第一个分享想法的人!

ECC

by affaan-m

10

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

242,21936,702JavaScript
AI 智能体ai-agentsanthropicclaude-code
查看详情
15

An agentic skills framework & software development methodology that works.

234,96620,863Shell
AI 智能体ai-agentsbrainstorming
查看详情

hermes-agent

by NousResearch

10

The agent that grows with you

234,43747,175Python
AI 智能体ai-agentsagent-orchestration
查看详情

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

185,94028,768JavaScript
AI 智能体ai-agentsanthropicclaude-code
查看详情

cc-switch

by farion1231

3

A cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Grok Build & Hermes Agent. Only official website: ccswitch.io

128,8688,826Rust
AI 智能体claude-codeai-tools
查看详情

claude-code

by anthropics

Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.

120,03119,897Shell
AI 智能体
查看详情

开发者还喜欢

基于喜欢此 Skill 的开发者投票和收藏

ECC

by affaan-m

10

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

242,21936,702JavaScript
AI 智能体ai-agentsanthropicclaude-code
查看详情
15

An agentic skills framework & software development methodology that works.

234,96620,863Shell
AI 智能体ai-agentsbrainstorming
查看详情

hermes-agent

by NousResearch

10

The agent that grows with you

234,43747,175Python
AI 智能体ai-agentsagent-orchestration
查看详情

n8n

by n8n-io

12

Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.

201,88160,308TypeScript
MCP 服务器apisai-tools
查看详情

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

185,94028,768JavaScript
AI 智能体ai-agentsanthropicclaude-code
查看详情

cc-switch

by farion1231

3

A cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Grok Build & Hermes Agent. Only official website: ccswitch.io

128,8688,826Rust
AI 智能体claude-codeai-tools
查看详情