cli-anything
Reference guide for CLI-Anything, which auto-generates production-ready agent-controllable CLI harnesses for any GUI application via a 7-phase.
What this skill does
# CLI-Anything Reference Auto-generates production-ready, agent-controllable CLI interfaces for any GUI application with accessible source code. A 7-phase pipeline maps GUI actions to backend APIs and emits Click-based harnesses with `--json` output mode, interactive REPL, and pytest test suites. **Source**: <https://github.com/HKUDS/CLI-Anything> **License**: MIT | **Python**: 3.10+ | **Tests**: 1,508 passing across 11 apps --- > **Self-Evolving Skill**: This skill improves through use. If instructions are wrong, parameters drifted, or a workaround was needed — fix this file immediately, don't defer. Only update for real, reproducible issues. ## Installation ### Method 1: Claude Code Marketplace (recommended) ``` /plugin marketplace add HKUDS/CLI-Anything /plugin install cli-anything ``` ### Method 2: Manual Claude Code ```bash git clone https://github.com/HKUDS/CLI-Anything.git cp -r CLI-Anything/cli-anything-plugin ~/.claude/plugins/cli-anything # Reload in Claude Code: /reload-plugins ``` ### Method 3: OpenCode — global ```bash git clone https://github.com/HKUDS/CLI-Anything.git cp CLI-Anything/opencode-commands/*.md ~/.config/opencode/commands/ cp CLI-Anything/cli-anything-plugin/HARNESS.md ~/.config/opencode/commands/ ``` ### Method 4: OpenCode — project-level ```bash cp CLI-Anything/opencode-commands/*.md .opencode/commands/ cp CLI-Anything/cli-anything-plugin/HARNESS.md .opencode/commands/ ``` --- ## Generating a New CLI Harness ```bash # From local source path /cli-anything ./gimp # From GitHub URL (auto-cloned) /cli-anything https://github.com/blender/blender # Claude Code variant (plugin command) /cli-anything ./inkscape # OpenCode variant /cli-anything ./libreoffice ``` --- ## All Commands | Command | Platform | Purpose | | ------------------------------------------------------ | ----------- | ------------------------------------- | | `/cli-anything <path-or-url>` | Claude Code | Build full 7-phase harness | | `/cli-anything:refine <path> [focus]` | Claude Code | Expand with gap analysis | | `/cli-anything:test <path-or-url>` | Claude Code | Run tests + update TEST.md | | `/cli-anything:validate <path-or-url>` | Claude Code | Verify HARNESS.md standards | | `/cli-anything:list [--path dir] [--depth n] [--json]` | Claude Code | Discover installed CLI-Anything tools | | `/cli-anything ./gimp` | OpenCode | Build harness | | `/cli-anything-refine ./gimp` | OpenCode | Refine harness | | `/cli-anything-test ./gimp` | OpenCode | Run tests | | `/cli-anything-validate ./gimp` | OpenCode | Validate standards | | `/cli-anything-list` | OpenCode | List installed tools | ### Refinement with Focus ```bash # Broad gap analysis /cli-anything:refine ./gimp # Targeted domain expansion /cli-anything:refine ./gimp "batch processing and filters" /cli-anything:refine ./blender "rendering pipeline and animations" /cli-anything:refine ./libreoffice "spreadsheet formulas and macros" ``` --- ## Installing a Generated CLI ```bash cd <software>/agent-harness pip install -e . # Verify PATH entry created which cli-anything-gimp cli-anything-gimp --help ``` --- ## Using Generated CLIs All generated CLIs share a consistent interface pattern. ### GIMP Example ```bash # Help cli-anything-gimp --help # Create project cli-anything-gimp project new --width 1920 --height 1080 -o poster.json # Machine-readable JSON output (for agent consumption) cli-anything-gimp --json layer add -n "Background" --type solid --color "#1a1a2e" # Interactive REPL cli-anything-gimp # Stateful workflow with --project flag cli-anything-gimp --project poster.json layer add -n "Logo" --type group cli-anything-gimp --project poster.json export render output.png --format png ``` ### LibreOffice Example ```bash # New document cli-anything-libreoffice document new -o report.json --type writer # Populate document cli-anything-libreoffice --project report.json writer add-heading -t "Q1 Report" --level 1 cli-anything-libreoffice --project report.json writer add-table --rows 4 --cols 3 # Export cli-anything-libreoffice --project report.json export render output.pdf -p pdf --overwrite # JSON output cli-anything-libreoffice --json document info --project report.json ``` ### Blender Example ```bash cli-anything-blender scene new -o scene.json cli-anything-blender --project scene.json object add --type mesh --mesh-type CUBE cli-anything-blender --project scene.json render image output.png --engine CYCLES cli-anything-blender --json scene info --project scene.json ``` ### Inkscape Example ```bash cli-anything-inkscape document new --width 800 --height 600 -o design.json cli-anything-inkscape --project design.json shape add rect --x 10 --y 10 --w 200 --h 100 cli-anything-inkscape --project design.json export render output.svg cli-anything-inkscape --json document info --project design.json ``` --- ## Output Modes Every command supports two output modes: | Mode | Usage | Audience | | ------------------------ | ---------------------------------------- | ------------------ | | Human-readable (default) | `cli-anything-gimp layer add ...` | Terminal / humans | | Machine-readable JSON | `cli-anything-gimp --json layer add ...` | AI agents, scripts | The `--json` flag is always the first argument, before any subcommands. --- ## Interactive REPL Enter by running the CLI with no subcommands: ```bash cli-anything-gimp cli-anything-blender cli-anything-libreoffice ``` REPL features: - 50-level undo/redo stack - Persistent command history - ANSI 256-color output (software-specific accent colors) - Table rendering, progress indicators - All subcommands available interactively --- ## Running Tests ```bash # Standard pytest run cd <software>/agent-harness python3 -m pytest cli_anything/<software>/tests/ -v # Validate the installed binary is invoked (not just the module) CLI_ANYTHING_FORCE_INSTALLED=1 python3 -m pytest cli_anything/<software>/tests/ -v -s # Coverage python3 -m pytest cli_anything/<software>/tests/ --cov=cli_anything/<software> # Per-software examples cd gimp/agent-harness && python3 -m pytest cli_anything/gimp/tests/ -v cd blender/agent-harness && python3 -m pytest cli_anything/blender/tests/ -v cd libreoffice/agent-harness && python3 -m pytest cli_anything/libreoffice/tests/ -v ``` --- ## Environment Variables | Variable | Purpose | | -------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------ | | `CLI_ANYTHING_FORCE_INSTALLED=1` | Forces tests to use the real installed binary (validates `which cli-anything-<software>` resolves correctly, not just module import) | --- ## Supported Applications (11 total, 1,508 tests) | Application | Category | Entry Point | | ----------- | ----------------- | -------------------------- | | GIMP | Creative (raster) | `cli-anything-gimp` | | Blender | Creative (3D) | `cli-anything-blender` | | Inkscape | Creative (vector) | `cli-anything-inkscape` | | Audacity | Media (audio) | `cli-anything-audacity` | | LibreOffice | Productivity | `cli-anything-libreoffice` | | OBS Studio | Media (streaming) | `cli-anything-obs-studio` | | Kdenlive | Media (vid
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.