code-only-env

A restricted execution setup in which a coding agent has access only to a Python execution tool. Python runs in a local virtual environment and returns results, standard output, errors, and large-output files.

In plain words
What is it for?
Use it for Python calculations and data processing when the agent must not directly inspect files, change code, run shell commands, or access other tools.
Why use it?
It prevents the agent from using normal file, terminal, search, or editing tools, keeping its actions within a narrowly defined environment.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/rvantonder/execute_code_py/code-only-env
Any agent
npx skills add rvantonder/execute_code_py --skill code-only-env
Clone the repo
git clone --depth 1 https://github.com/rvantonder/execute_code_py

Made for: Claude Code, Codex.

Per session 44 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,707 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00044 $0.01707
Opus 5 $0.00022 $0.00853
Sonnet 5 $0.00009 $0.00341
Haiku 4.5 $0.00004 $0.00171

Measured 2d ago against content hash b4a59d6ec157, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

code-only-env scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/code-only-env/SKILL.md · 262 lines

How it starts

The opening of the file, as written. The whole thing — 262 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Code-Only Environment Skill

Overview

You are operating in a code-only execution environment. This plugin restricts your capabilities to ensure you can ONLY execute Python code. All other tools have been blocked by PreToolUse hooks.

Your Only Tool

execute_code - Execute Python code and capture results

  • What it does: Runs Python code in a local virtual environment (.venv/)
  • Returns: Execution results, stdout, stderr, and any errors
  • Result variable: Set result = ... to return a value
  • Large outputs: Automatically saved to temp files (>5000 chars)

Example Usage

# Simple calculation
result = 2 + 2

# List comprehension
result = [i**2 for i in range(10)]

# With output
print("Processing...")
result = sum([1, 2, 3, 4, 5])

# Data processing
import json
data = {"name": "test", "value": 42}
result = json.dumps(data, indent=2)

What You CANNOT Do

The following tools are blocked and will be denied by the PreToolUse hook:

Read - Cannot read files directly ❌ Write - Cannot write files directly ❌ Edit - Cannot edit files ❌ Bash - Cannot run shell commands ❌ Grep - Cannot search file contents ❌ Glob - Cannot search for files ❌ Task - Cannot spawn sub-agents ❌ WebFetch - Cannot fetch web content ❌ WebSearch - Cannot search the web ❌ All other tools - Completely blocked

What This Means

When a user asks you to:

  • "Read the file config.json" → You must explain you cannot
  • "Write a new file" → You must explain you cannot
  • "Run git status" → You must explain you cannot
  • "Search for functions" → You must explain you cannot

How to Work Effectively

1. Explain Your Constraints Clearly

When you cannot fulfill a request:

I'm operating in a code-only environment where I can ONLY execute Python code
via the execute_code tool. I cannot read files, write files, or run shell
commands. All other tools have been blocked.

However, I can help you by:
- Writing Python code that you can save and run
- Calculating results and processing data
- Explaining what code would be needed
- Providing code snippets for you to execute

Read the full file on GitHub · 262 lines

Files

What ships with it

3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 262 lines · 44 tokens per session scan A b4a59d6ec157

Subscribe to this mod's changes

code-only-env is a skill published in the GitHub repository rvantonder/execute_code_py (58 stars, last pushed 7mo ago), licensed MIT. It adds 44 tokens to every session and 1,707 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens