sandbox

sandbox is a skill for Claude Code from Marcel-Bich/marcel-bich-claude-marketplace. It costs 189 tokens per session (1,570 once invoked), scanned A, original, MIT.

A workspace for writing pre-work when a product decision needs evidence, such as a measurement, mockup, or feasibility proof. It keeps this work separate from production code and versioned project files.

In plain words
What is it for?
Creating temporary research artifacts, mockups, measurements, and feasibility proofs, then promoting accepted artifacts into the sandbox folder.
Why use it?
It gives blocked clarification tasks a safe place to gather evidence without disturbing the implementation work.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: mentions subagents.

Part of the credo plugin — 26 skills, 15 commands, 6 hooks shipped together

Good fit Creating temporary research artifacts, mockups, measurements, and feasibility proofs, then promoting accepted artifacts into the sandbox folder.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/marcel-bich/marcel-bich-claude-marketplace/sandbox
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.

Any agent
npx skills add Marcel-Bich/marcel-bich-claude-marketplace --skill sandbox
Clone the repo
git clone --depth 1 https://github.com/Marcel-Bich/marcel-bich-claude-marketplace

Made for: Claude Code.

Or install credo, the plugin that ships this one along with the rest of its 26 skills, 15 commands, 6 hooks.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for sandbox

README.md
[![agentmods](https://agentmods.dev/badge/skills/marcel-bich/marcel-bich-claude-marketplace/sandbox/github.svg)](https://agentmods.dev/skills/marcel-bich/marcel-bich-claude-marketplace/sandbox)
Your own site
<a href="https://agentmods.dev/skills/marcel-bich/marcel-bich-claude-marketplace/sandbox"><img src="https://agentmods.dev/badge/skills/marcel-bich/marcel-bich-claude-marketplace/sandbox/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for sandbox

Your own site · 80×15
<a href="https://agentmods.dev/skills/marcel-bich/marcel-bich-claude-marketplace/sandbox"><img src="https://agentmods.dev/badge/skills/marcel-bich/marcel-bich-claude-marketplace/sandbox.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 189 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,570 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00189 $0.01570
Opus 5 $0.00095 $0.00785
Sonnet 5 $0.00038 $0.00314
Haiku 4.5 $0.00019 $0.00157

Measured 12d ago against content hash 7dcbf72d32c7, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

sandbox 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 12d 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.

plugins/credo/skills/sandbox/SKILL.md · 113 lines

How it starts

The opening of the file, as written. The whole thing — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.

sandbox - writing pre-work for clarify items

Clarify items often hang on a knowledge gap that is NOT a product decision: a missing measurement, a mockup, or a proof that something is feasible. The plan / clarify role (role-plan) must not touch production code and must not commit. The sandbox gives it a place to do writing pre-work OFF to the side - no production code, no git, no interference with the task / build agent - so the clarify question can be answered with evidence instead of guesswork.

Two folders

  • .credo/sandbox-tmp/ = UNVERSIONED work in progress. One folder per item: .credo/sandbox-tmp/<id>-<slug>/ with a README.md. This is where pre-work is built.
  • .credo/sandbox/ = the PROMOTION TARGET for accepted, version-worthy artifacts. It follows the credo folder-versioning policy: in an untracked repo it stays local; in a version-tracked repo (CREDO_VERSION_TRACKED=1) it IS versioned.
  • docs/ is NOT the promotion target. Extracting a promoted artifact into docs/ is a SEPARATE step the user prompts for explicitly; it never happens automatically.

exclude behavior (owned by credo-init)

The git-exclude is managed by credo-init in a single managed block in .git/info/exclude (see the credo-init policy). Do not write that block from anywhere else.

  • Default (untracked, CREDO_VERSION_TRACKED unset): all of .credo/** is excluded, so both sandbox-tmp/ and sandbox/ stay local. Nothing extra is needed.
  • Tracked (CREDO_VERSION_TRACKED=1): .credo/** is versioned EXCEPT the entries credo-init lists, which include .credo/sandbox-tmp/. So WIP stays local even in a tracked repo, while the promoted sandbox/ is versioned.

The helper scripts/credo-sandbox-init.sh never writes the managed block; it only WARNS when sandbox-tmp is not effectively git-excluded and points at /credo:setup to refresh it.

Reference search order

When resolving a sandbox reference for an item, ALWAYS look in this order:

  1. .credo/sandbox/<name>/ (promoted, accepted artifact) - use this if present.
  2. .credo/sandbox-tmp/<name>/ (WIP) - fall back to this.

Read the full file on GitHub · 113 lines

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. 12d ago First seen · 113 lines · 189 tokens per session scan A 7dcbf72d32c7

Subscribe to this mod's changes

sandbox is a skill published in the GitHub repository Marcel-Bich/marcel-bich-claude-marketplace (13 stars, last pushed 2d ago), licensed MIT. It adds 189 tokens to every session and 1,570 once invoked, about $0.0009 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.