Token导航 LogoToken导航

Agent Skill · 研究检索

ntm(恩特姆)

ntm 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

ntm 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为可写文件;许可证为 MIT,第三方扫描有未通过项,按安装量在研究检索类的 73,675 个 Skill 里排第 41,356。

总安装 1,176周安装 50GitHub Stars 63下载 412核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:ntm(恩特姆)
来源仓库:https://github.com/dicklesworthstone/agent_flywheel_clawdbot_skills_and_integrations
仓库路径:skills/ntm
安装命令:
npx skills add https://github.com/dicklesworthstone/agent_flywheel_clawdbot_skills_and_integrations --skill ntm
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/dicklesworthstone/agent_flywheel_clawdbot_skills_and_integrations --skill ntm

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

ntm 用于查找、检索和筛选相关信息,适合快速定位候选结果。适用于根据关键词或任务场景从来源中获取信息的场景。通过 npx skills add 命令安装,建议查看原始 README 掌握用法。使用前需确认权限与维护状态,避免触发联网或文件操作。ntm 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/ntm

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/dicklesworthstone/agent_flywheel_clawdbot_skil…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要可写文件,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

ntm 怎么安装?

在终端执行 npx skills add https://github.com/dicklesworthstone/agent_flywheel_clawdbot_skills_and_int…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

ntm 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

ntm 需要什么权限?

权限级别是可写文件。该 Skill 可能写入或修改本地文件,使用前需要确认目标目录和修改范围。

ntm 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

ntm 用的人多吗?

公开数据是累计安装 1,176 次,近一周安装 50 次,下载 412 次,GitHub 63 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

ntm 安全吗?

第三方扫描有未通过项,风险较高。除非你确认过源码,否则不建议直接安装到常用环境。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。

先想清楚的地方

  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
ntm

恩特姆 · 本页

1,17663可写文件MIT未通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

NTM — Named Tmux Manager

A Go CLI that transforms tmux into a multi-agent command center for orchestrating Claude Code, Codex, and Gemini agents in parallel. Spawn, manage, and coordinate AI agents across tiled panes with stunning TUI, automated context rotation, and deep integrations with the Agent Flywheel ecosystem.

Why This Exists

Managing multiple AI coding agents is painful:

  • Window chaos: Each agent needs its own terminal
  • Context switching: Jumping between windows breaks flow
  • No orchestration: Same prompt to multiple agents requires manual copy-paste
  • Session fragility: Disconnecting from SSH loses all agent sessions
  • No visibility: Hard to see agent status at a glance

NTM solves all of this with one session containing many agents, persistent across SSH disconnections.

Quick Start

# Install
curl -fsSL https://raw.githubusercontent.com/Dicklesworthstone/ntm/main/install.sh | bash

# Add shell integration
echo 'eval "$(ntm init zsh)"' >> ~/.zshrc && source ~/.zshrc

# Interactive tutorial
ntm tutorial

# Check dependencies
ntm deps -v

# Create multi-agent session
ntm spawn myproject --cc=2 --cod=1 --gmi=1

# Send prompt to all Claude agents
ntm send myproject --cc "Explore this codebase and summarize its architecture."

# Open command palette (or press F6 after `ntm bind`)
ntm palette myproject

Session Creation

Spawn Agents

ntm spawn myproject --cc=3 --cod=2 --gmi=1   # 3 Claude + 2 Codex + 1 Gemini
ntm quick myproject --template=go             # Full project scaffold + agents
ntm create myproject --panes=10               # Empty panes only
ntm spawn myproject --profiles=architect,implementer,tester

Agent Flags

FlagAgentCLI Command
--cc=NClaude Codeclaude
--cod=NCodex CLIcodex
--gmi=NGemini CLIgemini

Add More Agents

ntm add myproject --cc=2              # Add 2 more Claude agents
ntm add myproject --cod=1 --gmi=1     # Add mixed agents

Sending Prompts

ntm send myproject --cc "Implement user auth"     # To all Claude
ntm send myproject --cod "Write unit tests"       # To all Codex
ntm send myproject --gmi "Review and document"    # To all Gemini
ntm send myproject --all "Review current state"   # To ALL agents
ntm interrupt myproject                           # Ctrl+C to all

Session Navigation

CommandAliasDescription
ntm listlntList all tmux sessions
ntm attachrntAttach to session
ntm statussntShow pane details with agent counts
ntm viewvntUnzoom, tile layout, attach
ntm zoomzntZoom to specific pane
ntm dashboarddash, dInteractive visual dashboard
ntm killkntKill session (-f to force)

Command Palette

Fuzzy-searchable TUI with pre-configured prompts:

ntm palette myproject    # Open palette
ntm bind                 # Set up F6 keybinding
ntm bind --key=F5        # Use different key

Palette Features

  • Animated gradient banner with Catppuccin themes
  • Fuzzy search with live filtering
  • Pin/favorite commands (Ctrl+P / Ctrl+F)
  • Live preview pane with metadata
  • Quick select with numbers 1-9
  • Visual target selector (All/Claude/Codex/Gemini)

Palette Navigation

KeyAction
↑/↓ or j/kNavigate
1-9Quick select
EnterSelect command
EscBack / Quit
?Help overlay
Ctrl+PPin/unpin
Ctrl+FFavorite

Interactive Dashboard

ntm dashboard myproject   # Or: ntm dash myproject

Dashboard Features

  • Visual pane grid with color-coded agent cards
  • Live agent counts (Claude/Codex/Gemini/User)
  • Token velocity badges (tokens-per-minute)
  • Context usage indicators (green/yellow/orange/red)
  • Real-time refresh with r

Dashboard Navigation

KeyAction
↑/↓ or j/kNavigate panes
1-9Quick select
z or EnterZoom to pane
rRefresh
cView context
mOpen Agent Mail
qQuit

Output Capture

ntm copy myproject:1              # Copy specific pane
ntm copy myproject --all          # Copy all panes
ntm copy myproject --cc           # Copy Claude panes only
ntm copy myproject --pattern 'ERROR'  # Filter by regex
ntm copy myproject --code         # Extract code blocks only
ntm copy myproject --output out.txt   # Save to file
ntm save myproject -o ~/logs      # Save all outputs

Monitoring & Analysis

ntm activity myproject --watch    # Real-time activity
ntm health myproject              # Health status
ntm watch myproject --cc          # Stream output
ntm extract myproject --lang=go   # Extract code blocks
ntm diff myproject cc_1 cod_1     # Compare panes
ntm grep 'error' myproject -C 3   # Search with context
ntm analytics --days 7            # Session statistics
ntm locks myproject --all-agents  # File reservations

Activity States

StateIconDescription
WAITING●Idle, ready for work
GENERATING▶Producing output
THINKING◐Processing (no output yet)
ERROR✗Encountered error
STALLED◯Stopped unexpectedly

Checkpoints

ntm checkpoint save myproject -m "Before refactor"
ntm checkpoint list myproject
ntm checkpoint show myproject 20251210-143052
ntm checkpoint delete myproject 20251210-143052 -f

Context Window Rotation

NTM monitors context usage and auto-rotates agents before exhausting context.

How It Works

  1. Monitoring: Token usage estimated per agent
  2. Warning: Alert at 80% usage
  3. Compaction: Try /compact or summarization first
  4. Rotation: Fresh agent with handoff summary if needed

Context Indicators

ColorUsageStatus
Green< 40%Plenty of room
Yellow40-60%Comfortable
Orange60-80%Approaching threshold
Red> 80%Needs attention

Automatic Compaction Recovery

When context is compacted, NTM sends a recovery prompt:

[context_rotation.recovery]
enabled = true
prompt = "Reread AGENTS.md so it's still fresh in your mind. Use ultrathink."
include_bead_context = true   # Include project state from bv

Robot Mode (AI Automation)

Machine-readable JSON output for AI agents and automation.

State Inspection

ntm --robot-status              # Sessions, panes, agent states
ntm --robot-context=SESSION     # Context window usage
ntm --robot-snapshot            # Unified state: sessions + beads + mail
ntm --robot-tail=SESSION        # Recent pane output
ntm --robot-inspect-pane=SESS   # Detailed pane inspection
ntm --robot-files=SESSION       # File changes with attribution
ntm --robot-metrics=SESSION     # Session metrics
ntm --robot-plan                # bv execution plan
ntm --robot-dashboard           # Dashboard summary
ntm --robot-health              # Project health

Agent Control

ntm --robot-send=SESSION --msg="Fix auth" --type=claude
ntm --robot-spawn=SESSION --spawn-cc=2 --spawn-wait
ntm --robot-interrupt=SESSION
ntm --robot-assign=SESSION --assign-beads=br-1,br-2
ntm --robot-replay=SESSION --replay-id=ID

Bead Management

ntm --robot-bead-claim=BEAD_ID --bead-assignee=agent
ntm --robot-bead-create --bead-title="Fix bug" --bead-type=bug
ntm --robot-bead-show=BEAD_ID
ntm --robot-bead-close=BEAD_ID --bead-close-reason="Fixed"

CASS Integration

ntm --robot-cass-search="auth error" --cass-since=7d
ntm --robot-cass-context="how to implement auth"
ntm --robot-cass-status

Exit Codes

CodeMeaning
0Success
1Error
2Unavailable/Not implemented

Work Distribution

Integration with BV for intelligent work assignment:

ntm work triage               # Full triage with recommendations
ntm work triage --by-label    # Group by domain
ntm work triage --quick       # Quick wins only
ntm work alerts               # Stale issues, priority drift, cycles
ntm work search "JWT auth"    # Semantic search
ntm work impact src/api/*.go  # Impact analysis
ntm work next                 # Single best next action

Intelligent Assignment

ntm --robot-assign=myproject --assign-strategy=balanced  # Default
ntm --robot-assign=myproject --assign-strategy=speed     # Maximize throughput
ntm --robot-assign=myproject --assign-strategy=quality   # Best agent-task match
ntm --robot-assign=myproject --assign-strategy=dependency # Unblock downstream

Agent Capability Matrix

AgentBest At
ClaudeAnalysis, refactoring, documentation, architecture
CodexFeature implementation, bug fixes, quick tasks
GeminiDocumentation, analysis, features

Profiles & Personas

ntm profiles list                    # List profiles
ntm profiles show architect          # Show details
ntm spawn myproject --profiles=architect,implementer,tester
ntm spawn myproject --profile-set=backend-team

Built-in Profiles

architect, implementer, reviewer, tester, documenter

Agent Mail Integration

ntm mail send myproject --to GreenCastle "Review API changes"
ntm mail send myproject --all "Checkpoint: sync status"
ntm mail inbox myproject
ntm mail read myproject --agent BlueLake
ntm mail ack myproject 42

Pre-commit Guard

ntm hooks guard install    # Prevent conflicting commits
ntm hooks guard uninstall

Notifications

Multi-channel notifications for events:

[notifications]
enabled = true
events = ["agent.error", "agent.crashed", "agent.rate_limit"]

[notifications.desktop]
enabled = true

[notifications.webhook]
enabled = true
url = "https://hooks.slack.com/..."

Event Types

agent.error, agent.crashed, agent.rate_limit, rotation.needed, session.created, session.killed, health.degraded

Alerting System

Alert Types

TypeSeverityDescription
unhealthyHighAgent enters unhealthy state
degradedMediumAgent performance degrades
rate_limitedMediumAPI rate limit detected
restart_failedHighRestart attempt failed
max_restartsCriticalRestart limit exceeded
ntm --robot-alerts
ntm --robot-dismiss-alert=ALERT_ID

Command Hooks

# ~/.config/ntm/hooks.toml

[[command_hooks]]
event = "post-spawn"
command = "notify-send 'NTM' 'Agents spawned'"

[[command_hooks]]
event = "pre-send"
command = "echo \"$(date): $NTM_MESSAGE\" >> ~/.ntm-send.log"

Available Events

pre-spawn, post-spawn, pre-send, post-send, pre-add, post-add, pre-shutdown, post-shutdown

Safety System

Blocks dangerous commands from AI agents:

ntm safety status              # Protection status
ntm safety check "git reset --hard"
ntm safety install             # Install git wrapper + Claude hook
ntm safety uninstall

Protected Commands

PatternRiskAction
git reset --hardLoses uncommitted changesBlock
git push --forceOverwrites remote historyBlock
rm -rf /Catastrophic deletionBlock
DROP TABLEDatabase destructionBlock

Multi-Agent Strategies

Divide and Conquer

ntm send myproject --cc "design the database schema"
ntm send myproject --cod "implement the models"
ntm send myproject --gmi "write tests"

Competitive Comparison

ntm send myproject --all "implement a rate limiter"
ntm view myproject  # Compare side-by-side

Review Pipeline

ntm send myproject --cc "implement feature X"
ntm send myproject --cod "review Claude's code"
ntm send myproject --gmi "write tests for edge cases"

Configuration

ntm config init          # Create ~/.config/ntm/config.toml
ntm config show          # Show current config
ntm config project init  # Create .ntm/config.toml in project

Example Config

projects_base = "~/Developer"

[agents]
claude = 'claude --dangerously-skip-permissions'
codex = "codex --dangerously-bypass-approvals-and-sandbox"
gemini = "gemini --yolo"

[tmux]
default_panes = 10
palette_key = "F6"

[context_rotation]
enabled = true
warning_threshold = 0.80
rotate_threshold = 0.95

Environment Variables

VariableDescription
NTM_PROJECTS_BASEBase directory for projects
NTM_THEMEColor theme: auto, mocha, latte, nord, plain
NTM_ICONSIcon set: nerd, unicode, ascii
NTM_REDUCE_MOTIONDisable animations
NTM_PROFILEEnable performance profiling

Themes & Display

Color Themes

ThemeDescription
autoDetect light/dark
mochaDefault dark, warm
latteLight variant
nordArctic-inspired
plainNo color

Agent Colors

AgentColor
ClaudeMauve (Purple)
CodexBlue
GeminiYellow
UserGreen

Display Width Tiers

WidthBehavior
<120 colsStacked layout
120-199 colsList/detail split
200-239 colsWider gutters
240+ colsFull detail

Pane Naming Convention

Pattern: <project>__<agent>_<number>

  • myproject__cc_1 - First Claude
  • myproject__cod_2 - Second Codex
  • myproject__gmi_1 - First Gemini

Status indicators: C = Claude, X = Codex, G = Gemini, U = User

Shell Aliases

After eval "$(ntm init zsh)":

CategoryAliases
Agent Launchcc, cod, gmi
Sessioncnt, sat, qps
Agent Mgmtant, bp, int
Navigationrnt, lnt, snt, vnt, znt
Dashboarddash, d
Outputcpnt, svnt
Utilitiesncp, knt, cad

Installation

# One-liner (recommended)
curl -fsSL https://raw.githubusercontent.com/Dicklesworthstone/ntm/main/install.sh | bash

# Homebrew
brew install dicklesworthstone/tap/ntm

# Go install
go install github.com/Dicklesworthstone/ntm/cmd/ntm@latest

# Docker
docker pull ghcr.io/dicklesworthstone/ntm:latest

Upgrade

ntm upgrade              # Check and install updates
ntm upgrade --check      # Check only
ntm upgrade --yes        # Auto-confirm

Tmux Essentials

KeysAction
Ctrl+B, DDetach
Ctrl+B, [Scroll/copy mode
Ctrl+B, zToggle zoom
Ctrl+B, ArrowNavigate panes
F6Open NTM palette (after ntm bind)

Integration with Flywheel

ToolIntegration
Agent MailMessage routing, file reservations, pre-commit guard
BVWork distribution, triage, assignment strategies
CASSSearch past sessions via robot mode
CMProcedural memory for agent handoffs
DCGSafety system integration
UBSAuto-scanning on file changes

平台分布

Claude Code

30.85%
按下载量换算127

Gemini CLI

25.42%
按下载量换算105

Antigravity

18.49%
按下载量换算76

OpenCode

12.76%
按下载量换算53

Codex

7.8%
按下载量换算32

windsurf

3.42%
按下载量换算14

安全审计

Gen Agent Trust Hub

未通过

Socket

可疑

Snyk

未通过

权限和风险

可写文件

该 Skill 可能写入或修改本地文件,使用前需要确认目标目录和修改范围。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。

继续浏览同类 Skills