environment-discovery — Claude Code skill: explore before you act
The environment-discovery skill trains Claude Code to systematically explore unfamiliar environments before making changes. It checks file structure, data formats, available tools, and existing code—saving debugging time.
MIT ★ 1,125 updated 2026-10-27
npx skills add vstorm-co/pydantic-deepagents --skill environment-discovery -a claude-codeWhat environment-discovery does
The skill adds four exploration steps to Claude Code. First, examine directory structure and look for READMEs, Makefiles, and config files. Second, inspect data files by detecting type with file, previewing with head, and reading hex dumps for binary data. Third, check which compilers, interpreters, and libraries are present using which and pip list. Fourth, read existing code fully and understand the build system.
The core principles are straightforward: never guess file formats, always verify tools are installed, and don't apply cat to binary files. For large files (500 MB or more), plan ahead instead of loading everything at once.
What it helps with
- Understand an unfamiliar project structure before making changes
- Determine data file format without guessing
- Check which compilers and libraries are available in the environment
- Avoid errors caused by unknown build configurations
Example prompts for the agent
- Help me understand what's in this project and where to start
- Look at the files in this directory and tell me what they contain
- Check if Python, gcc, and the libraries I need are installed
How to install environment-discovery
1Skills CLI (Vercel)
npx skills add vstorm-co/pydantic-deepagents --skill environment-discovery -a claude-code2Manually: put the files in place
~/.claude/skills/environment-discovery/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/environment-discovery/
1Skills CLI (Vercel)
npx skills add vstorm-co/pydantic-deepagents --skill environment-discovery -a codex2Manually: put the files in place
.agents/skills/environment-discovery/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add vstorm-co/pydantic-deepagents --skill environment-discovery -a cursor2Manually: put the files in place
.cursor/skills/environment-discovery/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install vstorm-co/pydantic-deepagents apps/cli/skills/environment-discovery2Manually: put the files in place
.github/skills/environment-discovery/SKILL.md1Skills CLI (Vercel)
npx skills add vstorm-co/pydantic-deepagents --skill environment-discovery -a gemini-cli2Platform CLI
gemini skills install https://github.com/vstorm-co/pydantic-deepagents --path apps/cli/skills/environment-discoveryManually: put the files in place
.devin/skills/environment-discovery/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about environment-discovery
How do I install environment-discovery in Claude Code?
Run npx skills add vstorm-co/pydantic-deepagents --skill environment-discovery -a claude-code, then edit ~/.claude/skills/environment-discovery/SKILL.md.
What does environment-discovery do?
The skill trains the agent to systematically explore an environment before starting work: check file structure, verify data formats, find installed tools, and understand existing code. This prevents errors and assumptions.
When should I use this skill?
Use it when entering an unfamiliar project or environment. It's especially useful when working with unknown data formats, binary files, or checking for specific tool availability.
Source text
SKILL.md · MIT
--- name: environment-discovery description: "Systematic exploration of unknown environments before starting work" tags: [exploration, setup, benchmark] version: "1.0.0" --- # Environment Discovery When dropped into an unfamiliar environment, ALWAYS explore before acting. ## Step 1: Understand the workspace ``` ls -la /app/ # or the working directory find . -type f | head -50 ``` - What files exist? What are their sizes? - Are there READMEs, Makefiles, config files? - What languages/frameworks are involved? ## Step 2: Inspect data files Before writing any code that reads data, understand the format: - `file <filename>` — detect file type (binary, text, encoding) - `head -20 <file>` — first lines of text files - `xxd <file> | head -20` — hex dump for binary files - `wc -l <file>` — line count for text files - `stat <file>` — exact file size in bytes - `python3 -c "import struct; ..."` — parse binary headers ## Step 3: Check available tools ``` which python3 gcc g++ make cmake node npm cargo rustc java go pip list 2>/dev/null | head -20 ``` - What compilers/interpreters are installed? - What libraries are available? - What package managers can you use? ## Step 4: Read existing code If there are existing source files: - Read them FULLY before modifying - Understand the build system (Makefile, CMakeLists.txt, pyproject.toml) - Check for existing tests ## Key Principles - NEVER assume file formats — always inspect first - NEVER assume tools are installed — always check - A 500MB file is NOT a "small file" — plan for it - Binary files need byte-level inspection, not `cat` - Spend 30 seconds exploring to save 5 minutes debugging
© vstorm-co · MIT · vstorm-co/pydantic-deepagents