HKUDS/CLI-Anything

  CLI-Anything: Making ALL Software Agent-Native

Today's Software Serves Humans👨‍💻. Tomorrow's Users will be Agents🤖.
CLI-Anything: Bridging the Gap Between AI Agents and the World's Software

🌐 CLI-Hub: pip install cli-anything-hub then cli-hub install <name> — browse, install, and manage all community-built CLIs. Want to add your own? Open a PR — the hub updates instantly.

🎬 See Demos: Watch AI agents use generated CLIs plus preview, live preview, and trajectory loops to produce real artifacts — CAD builds, 3D scenes, diagrams, gameplay, subtitles, and more.

🙋 [Become a Contributor, or Request a CLI]: Join us! Sign up to build a new CLI harness — once reviewed and merged, you'll gain access as one of our community contributors! Wish CLI-Anything supported a specific software or service? Submit a wishlist request!

One Command Line: Make any software agent-ready for Pi, OpenClaw, nanobot, Cursor, Claude Code, etc.  中文文档 | 日本語ドキュメント | Deutsch

📰 News

Thanks to all invaluable efforts from the community! More updates continuously on the way everyday..

• 2026-05-30 🧭 Hermes skill proposed (#320), adding a CLI-Anything orchestration skill with installer scripts and HARNESS fallback guidance. 🗺️ ArcGIS Pro was proposed for the public registry (#318) as a Windows/ArcPy CLI for cartography, geoprocessing, feature editing, and live-Pro MCP workflows.

• 2026-05-27 🔧 CLI-Hub registry date updates now handle python -m pip install commands (#312), improving package-date detection for registry automation.

• 2026-05-23 📝 Obsidian Agent CLI was proposed for the public registry (#307), bringing a PyPI-installed Obsidian automation CLI with persistent agent memory workflows and a pinned skill link.

• 2026-05-21 🔒 Sketch CLI token-file handling was hardened against path traversal and symlink escapes (#304).

• 2026-05-20 📓 Joplin CLI was proposed (#300) with notebooks, notes, to-dos, tags, attachments, search/sync, import/export, server/E2EE helpers, full docs, packaged skill docs, and a 134-test validation baseline.

• 2026-05-20 🎛️ Rekordbox CLI merged (#252) with guarded SQLCipher write paths, backup-required forced writes, smoke coverage, and root skill sync. 📚 Calibre CLI merged (#223) with library/search/metadata/conversion/export workflows, 41 unit tests, real-Calibre E2E evidence, and root skill validation. 🧊 3MF CLI merged (#209) with mesh inspection, hole resizing, repair, comparison, and preserved triangle attributes. 🎙️ MiniMax CLI merged (#189) with chat/TTS workflows, JSON-safe model/voice listing, REPL quote handling, and smoke/E2E coverage. 🎮 UEAtelier joined the registry (#297) as an Unreal Editor MCP self-extension workbench with a Python CLI proxy.

• 2026-05-19 🛠️ Existing harnesses got a quality/security pass — Zoom downloads recordings from direct URLs (#294), Obsidian search now uses the Local REST API vendor content types (#289), LibreOffice headless conversion is more robust on macOS (#290), and XML/SVG/ODF/MLT/MusicXML/CSL parsing now routes untrusted input through defusedxml (#296).

• 2026-05-18 📈 README presentation refreshed with the Trendshift badge and centered project header polish (#285, #286), keeping the landing section focused on discovery and project momentum.

• 2026-05-17 🌐 CLI-Hub registry handling was hardened (#281) — registry entries are now copied before _source tagging, preventing cached or mocked registry data from being mutated in place.

• 2026-05-16 🔧 n8n received the REPL banner crash fix that later merged into main (#280), restoring the no-subcommand interactive startup path with regression coverage.

🤔 Why CLI?

CLI is the universal interface for both humans and AI agents:

• Structured & Composable - Text commands match LLM format and chain for complex workflows

• Lightweight & Universal - Minimal overhead, works across all systems without dependencies

• Self-Describing - --help flags provide automatic documentation agents can discover

• Proven Success - Claude Code runs thousands of real workflows through CLI daily

• Agent-First Design - Structured JSON output eliminates parsing complexity

• Deterministic & Reliable - Consistent results enable predictable agent behavior

🚀 Quick Start

CLI-Anything has two equally useful starting points:

pip install cli-anything-hub

# Browse what is available
cli-hub list
cli-hub search image

# Install and try a CLI immediately
cli-hub install gimp
cli-hub info gimp
cli-hub launch gimp

npx skills add HKUDS/CLI-Anything --skill cli-hub-meta-skill -g -y

Find appropriate CLI software in CLI-Hub and complete the task: ...

# Add the CLI-Anything marketplace
/plugin marketplace add HKUDS/CLI-Anything

# Install the cli-anything plugin from the marketplace
/plugin install cli-anything

# /cli-anything <software-path-or-repo>
# Generate a complete CLI for GIMP (all 7 phases)
/cli-anything ./gimp

# Broad refinement — agent analyzes gaps across all capabilities
/cli-anything:refine ./gimp

# Focused refinement — target a specific functionality area
/cli-anything:refine ./gimp "I want more CLIs on image batch processing and filters"

# Clone the repo
git clone https://github.com/HKUDS/CLI-Anything.git

# Copy plugin to Claude Code plugins directory
cp -r CLI-Anything/cli-anything-plugin ~/.claude/plugins/cli-anything

# Reload plugins
/reload-plugins

# Clone the repo
git clone https://github.com/HKUDS/CLI-Anything.git
cd CLI-Anything

# Install globally into Pi's extensions directory
bash .pi-extension/cli-anything/install.sh

bash .pi-extension/cli-anything/install.sh --uninstall

# Generate a complete CLI for GIMP (all 7 phases)
/cli-anything ./gimp

# Build from a GitHub repo
/cli-anything https://github.com/blender/blender

# Broad refinement — agent analyzes gaps across all capabilities
/cli-anything:refine ./gimp

# Focused refinement — target a specific functionality area
/cli-anything:refine ./gimp "batch processing and filters"

# Clone the repo
git clone https://github.com/HKUDS/CLI-Anything.git

# Global install (available in all projects)
cp CLI-Anything/opencode-commands/*.md ~/.config/opencode/commands/
cp CLI-Anything/cli-anything-plugin/HARNESS.md ~/.config/opencode/commands/

# Or project-level install
cp CLI-Anything/opencode-commands/*.md .opencode/commands/
cp CLI-Anything/cli-anything-plugin/HARNESS.md .opencode/commands/

# Generate a complete CLI for GIMP (all 7 phases)
/cli-anything ./gimp

# Build from a GitHub repo
/cli-anything https://github.com/blender/blender

# Broad refinement — agent analyzes gaps across all capabilities
/cli-anything-refine ./gimp

# Focused refinement — target a specific functionality area
/cli-anything-refine ./gimp "batch processing and filters"

/cli-anything ./gimp
/cli-anything:refine ./gimp "batch processing and filters"

git clone https://github.com/HKUDS/CLI-Anything.git
bash CLI-Anything/qoder-plugin/setup-qodercli.sh

/cli-anything ./gimp
/cli-anything:refine ./gimp "batch processing and filters"
/cli-anything:validate ./gimp

# Clone the repo
git clone https://github.com/HKUDS/CLI-Anything.git

# Install to the global skills folder
mkdir -p ~/.openclaw/skills/cli-anything
cp CLI-Anything/openclaw-skill/SKILL.md ~/.openclaw/skills/cli-anything/SKILL.md

git clone https://github.com/HKUDS/CLI-Anything.git
bash CLI-Anything/cursor-plugin/scripts/install.sh

.\CLI-Anything\cursor-plugin\scripts\install.ps1

/cli-anything ./gimp
/cli-anything-refine ./shotcut "picture-in-picture workflows"
/cli-anything-test ./libreoffice
/cli-anything-validate ./libreoffice
/cli-anything-list

npx skills add HKUDS/CLI-Anything --skill cli-hub-meta-skill -g -y
# or a per-app skill, e.g. cli-anything-blender
pip install cli-anything-hub

bash CLI-Anything/cursor-plugin/tests/test_install.sh

# Clone the repo
git clone https://github.com/HKUDS/CLI-Anything.git

# Install the skill
bash CLI-Anything/codex-skill/scripts/install.sh

.\CLI-Anything\codex-skill\scripts\install.ps1

Use CLI-Anything to build a harness for ./gimp
Use CLI-Anything to refine ./shotcut for picture-in-picture workflows
Use CLI-Anything to validate ./libreoffice
Use CLI-Anything to list generated harnesses under the current directory

bash CLI-Anything/codex-skill/tests/test_install.sh

# Clone the repo
git clone https://github.com/HKUDS/CLI-Anything.git

# Install the skill
bash CLI-Anything/hermes-skill/scripts/install.sh

.\CLI-Anything\hermes-skill\scripts\install.ps1

Use CLI-Anything to build a harness for ./gimp
Use CLI-Anything to refine ./shotcut for picture-in-picture workflows
Use CLI-Anything to validate ./libreoffice

# Clone the repo
git clone https://github.com/HKUDS/CLI-Anything.git

# Install the skill
bash CLI-Anything/reasonix-skill/scripts/install.sh

.\CLI-Anything\reasonix-skill\scripts\install.ps1

Use CLI-Anything to build a harness for ./gimp
Use CLI-Anything to refine ./shotcut for picture-in-picture workflows
Use CLI-Anything to validate ./libreoffice

git clone https://github.com/HKUDS/CLI-Anything.git
cd CLI-Anything
copilot plugin install ./cli-anything-plugin

/cli-anything ./gimp
/cli-anything:refine ./gimp "batch processing and filters"
/cli-anything:validate ./gimp

# Install to PATH
cd gimp/agent-harness && pip install -e .

# Use from anywhere
cli-anything-gimp --help
cli-anything-gimp project new --width 1920 --height 1080 -o poster.json
cli-anything-gimp --json layer add -n "Background" --type solid --color "#1a1a2e"

# Enter interactive REPL
cli-anything-gimp

🤖 Empower Your Agents with CLI-Hub

CLI-Hub lets agents autonomously discover, install, and use the CLIs they need.

npx skills add HKUDS/CLI-Anything --skill cli-hub-meta-skill -g -y

Also available on: ClawHub, SkillHub, SkillHub.cn

Then prompt:

Find appropriate CLI software in CLI-Hub and complete the task: ...

The meta-skill points agents to the live CLI-Hub catalog, where they can choose a CLI, install it, and read its own SKILL.md for task-specific usage.

💡 CLI-Anything's Vision: Building Agent-Native Software

• 🌐 Universal Access - Every software becomes instantly agent-controllable through structured CLI.

• 🔗 Seamless Integration - Agents control any application without APIs, GUI, rebuilding or complex wrappers.

• 🚀 Future-Ready Ecosystem - Transform human-designed software into agent-native tools with one command.

🔧 When to Use CLI-Anything

CLI-Anything's Key Features

The Agent-Software Gap

AI agents are great at reasoning but terrible at using real professional software. Current solutions are fragile UI automation, limited APIs, or dumbed-down reimplementations that miss 90% of functionality.

CLI-Anything's Solution: Transform any professional software into agent-native tools without losing capabilities.

🎯 What Can You Do with CLI-Anything?

✨ ⚙️ How CLI-Anything Works

🤖 SKILL.md Generation

Each generated CLI now has a canonical SKILL.md at skills/cli-anything-<software>/SKILL.md. This makes the current monorepo directly consumable by npx skills, while a packaged compatibility copy at cli_anything/<software>/skills/SKILL.md preserves installed-harness behavior.

What SKILL.md provides:

• YAML frontmatter with name and description for agent skill discovery
• Command groups with all available subcommands documented
• Usage examples for common workflows
• Agent-specific guidance for JSON output, error handling, and programmatic use

SKILL.md files are auto-generated during Phase 6.5 of the pipeline using skill_generator.py, which extracts metadata directly from the CLI's Click decorators, setup.py, and README. The generator now writes the canonical repo-root skill file and refreshes the package-local compatibility copy used by installed harnesses. Inside this repo, the REPL banner points agents to the canonical root skill path; after pip install, it falls back to the packaged copy.

🎬 Real-World Demos

AI agents using generated CLIs to produce complete, useful artifacts — no GUI needed.

FreeCAD — Curiosity Rover via Preview, Live Preview, and Trajectory

Harness: cli-anything-freecad | Preview Stack: preview + preview live + trajectory.json | Artifact: Agent-built Curiosity-style rover

An agent incrementally assembles a Curiosity-inspired rover while publishing real FreeCAD preview bundles, refreshing a live preview session, and recording command-to-preview history for later replay. The resulting demo shows the artifact evolving step by step before the final showcase.

_{README GIF generated from the full local demo video with a speed-adjusted, high-quality ffmpeg palette workflow.}

Blender — Orbital Relay Drone via Preview, Live Preview, and Trajectory

Harness: cli-anything-blender | Preview Stack: preview + preview live + trajectory.json | Artifact: Agent-built orbital relay drone

An agent uses the Blender harness to grow a hard-surface orbital relay drone under a real preview loop: each stage pushes new render-backed bundles, the live session tracks the current head, and the trajectory ties every command to the matching visual state. The demo finishes with the completed scene ready for a polished turntable.

_{README GIF generated from the full local demo video with a speed-adjusted, high-quality ffmpeg palette workflow.}

Draw.io — HTTPS Handshake Diagram

Harness: cli-anything-drawio | Time: ~4 min | Artifact: .drawio + .png

An agent creates a full HTTPS connection lifecycle diagram from scratch — TCP three-way handshake, TLS negotiation, encrypted data exchange, and TCP four-way termination — entirely through CLI commands.

Contributed by @zhangxilong-43

Slay the Spire II — Game Automation

Harness: cli-anything-slay-the-spire-ii | Artifact: Automated gameplay session

An agent plays through a Slay the Spire II run using the CLI harness — reading game state, selecting cards, choosing paths, and making strategic decisions in real-time.

Contributed by @TianyuFan0504

VideoCaptioner — Auto-Generated Subtitles

Harness: cli-anything-videocaptioner | Artifact: Captioned video frames

An agent uses the VideoCaptioner CLI to automatically generate and overlay styled subtitles onto video content, with bilingual text rendering and customizable formatting.

Sub A	Sub B

Contributed by @WEIFENG2333

ArcGIS Pro — Live Agent Cartography via the MCP Bridge

Harness: arcgis-pro (registered in public_registry.json) | Mode: live-Pro MCP bridge | Artifact: Agent-driven map navigation in a running ArcGIS Pro session

An agent drives a live, open ArcGIS Pro session through an MCP bridge — reading the project, zooming the map to feature layers, running geoprocessing, and exporting layouts — while each step executes inside Pro as you watch. ArcGIS Pro is Esri's commercial GIS desktop (Windows-only, licensed), so this wraps its official ArcPy / ArcGIS Pro SDK rather than being generated from source — the ArcGIS Pro counterpart to the QGIS harness.

_{README GIF generated from the full local demo video with a high-quality ffmpeg palette workflow.}

Contributed by @Jasper0122

More CLI demos coming soon.

🎬 Demonstrations

🎯 General-Purpose

CLI-Anything works on any software with a codebase — no domain restrictions or architectural limitations.

🏭 Professional-Grade Testing

Tested across 18 diverse, complex applications spanning creative, productivity, communication, diagramming, AI image generation, AI content generation, network ad blocking, local LLM inference, native debugging, and graphics profiling domains previously inaccessible to AI agents.

🎨 Diverse Domain Coverage

From creative workflows (image editing, 3D modeling, vector graphics) to production tools (audio, office, live streaming, video editing).

✅ Full CLI Generation

Each application received complete, production-ready CLI interfaces — not demos, but comprehensive tool access preserving full capabilities.

100% pass rate across all 2,461 tests — 1,732 unit tests + 579 end-to-end tests + 19 Node.js tests.

📊 Test Results

Each CLI harness undergoes rigorous multi-layered testing to ensure production reliability:

================================ Test Summary ================================
gimp          107 passed  ✅   (64 unit + 43 e2e)
blender       208 passed  ✅   (150 unit + 58 e2e)
inkscape      202 passed  ✅   (148 unit + 54 e2e)
audacity      161 passed  ✅   (107 unit + 54 e2e)
wavetone       30 passed  ✅   (24 unit + 6 default e2e, 2 backend-gated e2e skipped; 32 passed opt-in)
libreoffice   158 passed  ✅   (89 unit + 69 e2e)
mubu           96 passed  ✅   (85 unit + 11 e2e)
obs-studio    153 passed  ✅   (116 unit + 37 e2e)
nslogger      139 passed  ✅   (97 unit + 42 e2e)
kdenlive      155 passed  ✅   (111 unit + 44 e2e)
shotcut       154 passed  ✅   (110 unit + 44 e2e)
zoom           22 passed  ✅   (22 unit + 0 e2e)
drawio        138 passed  ✅   (116 unit + 22 e2e)
eth2-quickstart 18 passed ✅   (18 unit + 3 e2e skipped)
mermaid        10 passed  ✅   (5 unit + 5 e2e)
anygen         50 passed  ✅   (40 unit + 10 e2e)
notebooklm     21 passed  ✅   (21 unit + 0 e2e)
comfyui        73 passed  ✅   (63 unit + 10 e2e)
adguardhome    36 passed  ✅   (24 unit + 12 e2e)
ollama         98 passed  ✅   (87 unit + 11 e2e)
sketch         19 passed  ✅   (19 jest, Node.js)
renderdoc      59 passed  ✅   (45 unit + 14 e2e)
cloudcompare   88 passed  ✅   (49 unit + 39 e2e)
openscreen    101 passed  ✅   (78 unit + 23 e2e)
lldb           27 passed  ✅   (23 unit + 4 e2e)
nsight-graphics 51 passed ✅   (46 unit/CLI + 5 local e2e)
unrealinsights 50 passed  ✅   (49 unit + 1 e2e, 9 backend-gated e2e skipped)
cloudanalyzer  14 passed  ✅   (7 unit + 7 e2e)
3mf            50 passed  ✅   (50 unit)
joplin        134 passed  ✅   (107 unit + 27 e2e, 1 skipped on Windows)
──────────────────────────────────────────────────────────────────────────────
TOTAL        2,464 passed  ✅   100% pass rate

🏗️ CLI-Anything's Architecture

🎯 Core Design Principles

1. Authentic Software Integration — The CLI generates valid project files (ODF, MLT XML, SVG) and delegates to real applications for rendering. We build structured interfaces TO software, not replacements.

2. Flexible Interaction Models — Every CLI operates in dual modes: stateful REPL for interactive agent sessions + subcommand interface for scripting/pipelines. Run bare command → enter REPL mode.

3. Consistent User Experience — All generated CLIs share unified REPL interface (repl_skin.py) with branded banners, styled prompts, command history, progress indicators, and standardized formatting.

4. Agent-Native Design — Built-in --json flag on every command delivers structured data for machine consumption, while human-readable tables serve interactive use. Agents discover capabilities via standard --help and which commands.

5. Zero Compromise Dependencies — Real software is a hard requirement — no fallbacks, no graceful degradation. Tests fail (not skip) when backends are missing, ensuring authentic functionality.

📂 Project Structure

cli-anything/
├── 📄 README.md                          # You are here
├── 📁 assets/                            # Images and media
│   ├── icon.png                          # Project icon
│   └── teaser.png                        # Teaser figure
│
├── 🔌 cli-anything-plugin/               # The Claude Code plugin
│   ├── HARNESS.md                        # Methodology SOP (source of truth)
│   ├── README.md                         # Plugin documentation
│   ├── QUICKSTART.md                     # 5-minute getting started
│   ├── PUBLISHING.md                     # Distribution guide
│   ├── repl_skin.py                      # Unified REPL interface
│   ├── commands/                         # Plugin command definitions
│   │   ├── cli-anything.md               # Main build command
│   │   ├── refine.md                     # Expand existing harness coverage
│   │   ├── test.md                       # Test runner
│   │   └── validate.md                   # Standards validation
│   └── scripts/
│       └── setup-cli-anything.sh         # Setup script
│
├── 🖱️ cursor-plugin/                     # Cursor Desktop generator plugin
│   ├── .cursor-plugin/plugin.json        # Cursor plugin manifest
│   ├── commands/                         # /cli-anything slash commands
│   ├── skills/                           # Supporting generator skill
│   ├── rules/                            # Phase-gate guidance
│   ├── scripts/                          # Bash/PowerShell installers + vendored helpers
│   └── tests/                            # Installer resource-sync regression tests
├── 🤖 codex-skill/                      # Self-contained Codex skill installer
│   ├── SKILL.md                         # Codex workflow entry point
│   ├── agents/                          # Codex UI metadata
│   ├── scripts/                         # Bash and PowerShell installers
│   └── tests/                           # Installer resource-sync regression test
├── 🧭 hermes-skill/                     # Hermes Agent skill entry point
├── 🧠 reasonix-skill/                   # Reasonix skill entry point
├── 🎨 gimp/agent-harness/               # GIMP CLI (107 tests)
├── 🧊 blender/agent-harness/            # Blender CLI (208 tests)
├── ✏️ inkscape/agent-harness/            # Inkscape CLI (202 tests)
├── 🎵 audacity/agent-harness/           # Audacity CLI (161 tests)
├── wavetone/agent-harness/              # WaveTone CLI (32 tests: 30 default + 2 backend-gated e2e)
├── 🌐 browser/agent-harness/            # Browser CLI (DOMShell MCP, new)
├── 🌐 web-yu-pri/agent-harness/         # Japan Post Web Yu-pri CLI (new)
├── 📄 libreoffice/agent-harness/        # LibreOffice CLI (158 tests)
├── 🧹 openrefine/agent-harness/         # OpenRefine CLI (76 tests: 64 unit + 12 real backend e2e)
├── 📧 mailchimp/agent-harness/          # Mailchimp Marketing API CLI (303 commands, 36 unit tests)
├── 📚 zotero/agent-harness/             # Zotero CLI (new, write import support)
├── 📖 calibre/agent-harness/            # Calibre CLI (58 tests: 38 unit + 20 E2E)
├── 📓 joplin/agent-harness/             # Joplin CLI (134 tests: 107 unit + 27 e2e)
├── 📝 mubu/agent-harness/               # Mubu CLI (96 tests)
├── 📹 obs-studio/agent-harness/         # OBS Studio CLI (153 tests)
├── 📱 nslogger/agent-harness/           # NSLogger CLI (139 tests)
├── 🎞️ kdenlive/agent-harness/           # Kdenlive CLI (155 tests)
├── 🎬 shotcut/agent-harness/            # Shotcut CLI (154 tests)
├── 📞 zoom/agent-harness/               # Zoom CLI (22 tests)
├── 🎵 musescore/agent-harness/          # MuseScore CLI (56 tests)
├── 📐 drawio/agent-harness/             # Draw.io CLI (138 tests)
├── 🧪 eez-studio/agent-harness/         # EEZ Studio CLI (project, LVGL, SCPI automation)
├── ⛓️ eth2-quickstart/agent-harness/    # ETH2 QuickStart CLI (18 unit, 3 e2e skipped)
├── 🧜 mermaid/agent-harness/            # Mermaid Live Editor CLI (10 tests)
├── ✨ anygen/agent-harness/             # AnyGen CLI (50 tests)
├── 🖼️ comfyui/agent-harness/            # ComfyUI CLI (73 tests)
├── 🧠 notebooklm/agent-harness/         # NotebookLM CLI (experimental, 21 tests)
├── 🧩 dify-workflow/agent-harness/      # Dify Workflow CLI wrapper (11 tests)
├── 🛡️ adguardhome/agent-harness/       # AdGuard Home CLI (36 tests)
├── 🦙 ollama/agent-harness/             # Ollama CLI (98 tests)
├── 🎮 godot/agent-harness/              # Godot Engine CLI (24 tests)
├── 📦 sbox/agent-harness/               # s&box CLI (244 tests: 157 unit + 17 orchestrator + 50 e2e + 20 exit-code)
├── 🎨 sketch/agent-harness/             # Sketch CLI (19 tests, Node.js)
├── 🔬 renderdoc/agent-harness/          # RenderDoc CLI (59 tests)
├── 🟩 nsight-graphics/agent-harness/    # Nsight Graphics CLI (51 tests)
├── 🐞 lldb/agent-harness/               # LLDB CLI (27 tests)
├── 📈 unrealinsights/agent-harness/     # Unreal Insights CLI (50 tests)
├── 🎬 videocaptioner/agent-harness/     # VideoCaptioner CLI (26 tests)
├── 🎬 openscreen/agent-harness/         # Openscreen CLI — screen recording editor (101 tests)
├── ☁️ cloudcompare/agent-harness/       # CloudCompare CLI (88 tests)
├── 🔍 exa/agent-harness/               # Exa CLI (40 tests)
└── ⛅ cloudanalyzer/agent-harness/      # CloudAnalyzer CLI (14 tests)
└── 🔩 3MF/agent-harness/               # 3MF Mesh Editor CLI (50+ tests)

Each agent-harness/ contains an installable Python package under cli_anything.<software>/ with Click CLI, core modules, utils (including repl_skin.py and backend wrapper), and comprehensive tests.

🎯 Plugin Commands

Examples

# Build a complete CLI for GIMP from local source
/cli-anything /home/user/gimp

# Build from a GitHub repo
/cli-anything https://github.com/blender/blender

# Refine an existing harness — broad gap analysis
/cli-anything:refine /home/user/gimp

# Refine with a specific focus area
/cli-anything:refine /home/user/shotcut "vid-in-vid and picture-in-picture compositing"

# Run tests and update TEST.md
/cli-anything:test /home/user/inkscape

# Validate against HARNESS.md standards
/cli-anything:validate /home/user/audacity

🎮 Demo: Using a Generated CLI

Here's what an agent can do with cli-anything-libreoffice:

# Create a new Writer document
$ cli-anything-libreoffice document new -o report.json --type writer
✓ Created Writer document: report.json

# Add content
$ cli-anything-libreoffice --project report.json writer add-heading -t "Q1 Report" --level 1
✓ Added heading: "Q1 Report"

$ cli-anything-libreoffice --project report.json writer add-table --rows 4 --cols 3
✓ Added 4×3 table

# Export to real PDF via LibreOffice headless
$ cli-anything-libreoffice --project report.json export render output.pdf -p pdf --overwrite
✓ Exported: output.pdf (42,831 bytes) via libreoffice-headless

# JSON mode for agent consumption
$ cli-anything-libreoffice --json document info --project report.json
{
  "name": "Q1 Report",
  "type": "writer",
  "pages": 1,
  "elements": 2,
  "modified": true
}

REPL Mode

$ cli-anything-blender
╔══════════════════════════════════════════╗
║       cli-anything-blender v1.0.0       ║
║     Blender CLI for AI Agents           ║
╚══════════════════════════════════════════╝

blender> scene new --name ProductShot
✓ Created scene: ProductShot

blender[ProductShot]> object add-mesh --type cube --location 0 0 1
✓ Added mesh: Cube at (0, 0, 1)

blender[ProductShot]*> render execute --output render.png --engine CYCLES
✓ Rendered: render.png (1920×1080, 2.3 MB) via blender --background

blender[ProductShot]> exit
Goodbye! 👋

📖 The Standard Playbook: HARNESS.md

HARNESS.md is our definitive SOP for making any software agent-accessible via automated CLI generation.

It encodes proven patterns and methodologies refined through automated generation processes.

The playbook distills key insights from successfully building all 18 diverse, production-ready harnesses.

Critical Lessons

See the full methodology: cli-anything-plugin/HARNESS.md

📦 Installation & Usage

For CLI-Hub Users

# Install the package manager
pip install cli-anything-hub

# Browse, search, inspect, and install CLIs
cli-hub list
cli-hub search <query>
cli-hub info <name>
cli-hub install <name>

# Manage installed CLIs
cli-hub update <name>
cli-hub uninstall <name>
cli-hub launch <name> [args...]

For Plugin Users (Claude Code)

# Add marketplace & install (recommended)
/plugin marketplace add HKUDS/CLI-Anything
/plugin install cli-anything

# Build a CLI for any software with a codebase
/cli-anything <software-name>

For Generated CLIs

# Install any generated CLI
cd <software>/agent-harness
pip install -e .

# Verify
which cli-anything-<software>

# Use
cli-anything-<software> --help
cli-anything-<software>                    # enters REPL
cli-anything-<software> --json <command>   # JSON output for agents

Running Tests

# Run tests for a specific CLI
cd <software>/agent-harness
python3 -m pytest cli_anything/<software>/tests/ -v

# Force-installed mode (recommended for validation)
CLI_ANYTHING_FORCE_INSTALLED=1 python3 -m pytest cli_anything/<software>/tests/ -v -s

🤝 Contributing

We welcome contributions! CLI-Anything is designed to be extensible:

• New software targets — Use the plugin to generate a CLI for any software with a codebase, then submit your harness via cli-anything-plugin/PUBLISHING.md.
• Methodology improvements — PRs to HARNESS.md that encode new lessons learned
• Plugin enhancements — New commands, phase improvements, better validation
• Test coverage — More E2E scenarios, edge cases, workflow tests

Limitations

• Requires strong foundation models — CLI-Anything relies on frontier-class models (e.g., Claude Opus 4.6, Claude Sonnet 4.6, GPT-5.4) for reliable harness generation. Weaker or smaller models may produce incomplete or incorrect CLIs that require significant manual correction.
• Relies on available source code — The 7-phase pipeline analyzes and generates from source code. When the target software only provides compiled binaries that require decompilation, harness quality and coverage will degrade substantially.
• May require iterative refinement — A single /cli-anything run may not fully cover all capabilities. Running /refine one or more times is often needed to push the CLI's performance and coverage to production quality.

Roadmap

• Support for more application categories (CAD, DAW, IDE, EDA, scientific tools)
• Benchmark suite for agent task completion rates
• Community-contributed CLI harnesses for internal/custom software
• Integration with additional agent frameworks beyond Claude Code
• Support packaging APIs for closed-source software and web services into CLIs
• Produce SKILL.md alongside the CLI for agent skill discovery and orchestration

📖 Documentation

Each generated harness also includes:

• <SOFTWARE>.md — Architecture SOP specific to that application
• tests/TEST.md — Test plan and results documentation

⭐ Star History

If CLI-Anything helps make your software Agent-native, give us a star! ⭐

📚 Citation

If you find CLI-Anything useful, please cite our technical report:

@misc{yang2026clianythingagentnativecomputeruse,
      title={CLI-Anything: Towards Agent-Native Computer Use}, 
      author={Yuhao Yang and Tianyu Fan and Chao Huang},
      year={2026},
      eprint={2606.03854},
      archivePrefix={arXiv},
      primaryClass={cs.HC},
      url={https://arxiv.org/abs/2606.03854}, 
}

📄 License

Apache License 2.0 — free to use, modify, and distribute.

CLI-Anything — Make any software with a codebase Agent-native.

_{A methodology for the age of AI agents | 18 professional software demos | 2,280 passing tests}

Thanks for visiting ✨ CLI-Anything!