A safety setting for a research project that requires confirmation before risky file or version-control actions. It covers actions such as deleting files, resetting Git history, force-pushing, or replacing important manuscript and output files.

Use it to turn the confirmation requirement on or off for destructive changes involving the paper, analysis outputs, submission files, or preregistration.

It helps prevent accidental loss or overwriting of work, especially near a submission deadline or during unattended work.

careful — Skill for Claude Code and Codex by matthewdigiuseppe: A safety setting for a research project that requires…

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/matthewdigiuseppe/mstack/careful

Any agent

npx skills add matthewdigiuseppe/MStack --skill careful

Clone the repo

git clone --depth 1 https://github.com/matthewdigiuseppe/MStack

Made for: Claude Code, Codex.

Or install mstack, the plugin that ships this one along with the rest of its 38 skills, 1 hook.

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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/matthewdigiuseppe/mstack/careful.svg)](https://agentmods.dev/skills/matthewdigiuseppe/mstack/careful)

Your own site

<a href="https://agentmods.dev/skills/matthewdigiuseppe/mstack/careful"><img src="https://agentmods.dev/badge/skills/matthewdigiuseppe/mstack/careful.svg" alt="Measured on agentmods" height="20"></a>

Per session 59 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 640 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.

Model	Per session	Once invoked
Fable 5	$0.00059	$0.00640
Opus 5	$0.00030	$0.00320
Sonnet 5	$0.00012	$0.00128
Haiku 4.5	$0.00006	$0.00064

Measured 5d ago against content hash 9b2cd0a9a7c0, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

careful 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 5d 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/careful/SKILL.md · 51 lines

How it starts

The opening of the file, as written. The whole thing — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/mstack:careful

Stage: power Voice: safety

When to invoke

Near a deadline. After a major edit you don't want clobbered. Before letting Claude work autonomously on a paper that's about to be submitted.

How it's enforced

This is not a promise — it's a mechanism. MStack ships a PreToolUse hook (hooks/mstack-guard.py) that runs before every Write, Edit, and Bash call. When .mstack/safety.yaml has careful: true, the hook turns the actions below into an explicit confirmation prompt, so they cannot run silently even in a fresh session that never loaded this skill.

Procedure

Read or create .mstack/safety.yaml in the current paper folder. If missing, create with default careful: false.
Toggle. Flip careful to true (or false if $ARGUMENTS is off). Write the flag as a top-level careful: true|false line — the hook parses this file.
What the hook intercepts while careful: true (each becomes a confirm-first prompt):
- rm, git reset --hard, git checkout --, git clean -f, force push, sed -i, truncate.
- mv touching paper/, output/, submission/, or prereg/.
- Write that would replace an existing non-empty file under paper/, output/, submission/, prereg/, or any existing file of 100+ lines.
- Reads, greps, targeted Edits, and writes to .mstack/, code/, and data/ (non-raw) proceed normally. (data/raw/ is protected by the hook at all times, independent of this toggle.)
Print the current state to the user: careful mode is now <on|off>, enforced by the MStack guard hook.

Outputs

.mstack/safety.yaml updated.
Summary block: current state.

Anti-patterns to refuse

Bypassing while careful is on. If the user confirms a prompted command, that approval covers that command only — don't flip the flag off to avoid future prompts unless the user runs /mstack:careful off.

When to call other skills

Pair with /mstack:freeze for stricter lockdown — that becomes /mstack:guard.

Read the full file on GitHub · 51 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.

5d ago First seen · 51 lines · 59 tokens per session scan A 9b2cd0a9a7c0

Subscribe to this mod's changes

careful is a skill published in the GitHub repository matthewdigiuseppe/MStack (14 stars, last pushed 8d ago), licensed MIT. It adds 59 tokens to every session and 640 once invoked, about $0.0003 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.

Git and workflow Security

systematic-debugging

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

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens

chat-perf

Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.

microsoft/vscode · 51 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 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