SkillAnything Skill Generator

What it does

When your agent uses it

• Creating a skill for a CLI tool or API that has none yet
• Turning an existing tool into an agent-ready skill with tests
• Packaging one skill for more than one agent platform
• Reviewing each stage of skill generation before the next begins

Example prompts

• “Create a skill for the jq CLI tool.”
• “Generate a skill for the Stripe API and pause after each phase so I can review it.”
• “Run only the analyze phase on the httpie CLI and show me analysis.json.”

Requirements

• Python to run the pipeline scripts
• A target to analyze, such as a CLI tool, API or library

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

1. 1Test cases with assertions → graded by agents/grader.md
2. 2Benchmark comparing with-skill vs baseline → benchmark.json
3. 3Description optimization with train/test split → prevents overfitting
4. 4Interactive viewer via eval-viewer/generate_review.py

What it can do on your machine

Read from SKILL.md and the folder at commit 0f7d527. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Ships 1 file in scripts/, which the agent can run.

  Shell commands in SKILL.md call:

  • python
  • pip

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md. Its commands use pip, which can reach the network depending on how they are called.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

SkillAnything Skill Generator loads about 1.9k tokens when it runs, and up to ~13k if it reads all its reference files. Until then it costs about 188 tokens; SKILL.md has 544 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~188

When it runs · the whole SKILL.md, loaded when a task matches

~1.9k

With references · SKILL.md plus every file in references/, read only if the agent opens them

~13k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from AgentSkillOS/SkillAnything at commit 0f7d527, republished under its MIT licence (© AgentSkillOS). 544 words, ~1,867 tokens.

name

skill-anything

description

Automatically generate production-ready Skills for any software, API, CLI tool, library, workflow, or service. Use this skill whenever the user wants to create a skill from scratch for a target application, convert an existing tool into an agent-native skill, generate skills for multiple platforms (Claude Code, OpenClaw, Codex), or automate the full skill creation pipeline including analysis, design, implementation, testing, optimization, and multi-platform packaging. Also use when the user mentions "skill-anything", "generate a skill for", "make a skill from", "skillify", or wants to turn any software into an agent-ready skill. Even if they just say "create a skill for X" where X is any tool or API, this skill should trigger.

Give SkillAnything a target and it handles everything:
- "Create a skill for the jq CLI tool"
- "Generate a skill for the Stripe API"
- "Turn this workflow into a multi-platform skill"

Phase 1: Analyze    → Detect target type, extract capabilities     → analysis.json
Phase 2: Design     → Map capabilities to skill architecture       → architecture.json
Phase 3: Implement  → Generate SKILL.md + scripts + references     → complete skill directory
Phase 4: Test Plan  → Auto-generate eval cases + trigger queries   → evals.json
Phase 5: Evaluate   → Benchmark with/without skill, grade results  → benchmark.json
Phase 6: Optimize   → Improve description via train/test loop      → optimized SKILL.md
Phase 7: Package    → Multi-platform distribution packages         → dist/

Target: "the httpie CLI tool"
→ Analyzes httpie --help output, designs command structure, generates skill,
  creates tests, benchmarks, optimizes, packages for 4 platforms

python -m scripts.analyze_target --target "jq" --output analysis.json
python -m scripts.design_skill --analysis analysis.json --output architecture.json
python -m scripts.init_skill my-skill --template cli --output ./out
python -m scripts.generate_tests --analysis analysis.json --skill-path ./out/my-skill
python -m scripts.run_eval --eval-set evals.json --skill-path ./out/my-skill
python -m scripts.run_loop --eval-set trigger-evals.json --skill-path ./out/my-skill --model <model>
python -m scripts.package_multiplatform ./out/my-skill --platforms claude-code,openclaw,codex

© AgentSkillOS, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 57 other files (scripts, references, assets) in the repository root of AgentSkillOS/SkillAnything.

• SKILL.md
• .github/profile/README.md
• .gitignore
• LICENSE
• METHODOLOGY.md
• NOTICE
• README.md
• agents/analyzer.md
• agents/comparator.md
• agents/designer.md
• agents/grader.md
• agents/implementer.md
• agents/optimizer.md
• agents/packager.md
• assets/eval_review.html
• config.yaml
• eval-viewer
• … and 41 more

Open the folder on GitHubat commit 0f7d527