careful
Enable opt-in warnings or denials for risky Bash commands in the current Claude Code session.
Skill by puukis
Chalk Labs
Need help deploying careful?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/puukis-careful
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/careful
Tools it pre-approves 2
From allowed-tools in SKILL.md: the agent may use these without asking you each time.
Bash,AskUserQuestion
What's inside
- Skill name
careful- Description
- Enable opt-in warnings or denials for risky Bash commands in the current Claude Code session.
- License
- MIT
- Allowed tools
- Bash,, AskUserQuestion
- Instructions
- 24 lines in SKILL.md · view on GitHub
Before installing
- Pre-approves some shell commands
Files (1)
- SKILL.md
SKILL.md (first lines)
# Careful Use when the user says be careful, production mode, safety mode, shared environment, or asks to warn before destructive commands. ## Process 1. If the user explicitly asks for strict mode, run `lstack safety strict`. 2. Otherwise run `lstack safety careful`. 3. Run `lstack safety status` when useful to confirm the mode. 4. Explain the active behavior briefly. ## Modes - `careful`: asks before risky Bash commands when hook ask is supported. - `strict`: denies risky Bash commands. - `off`: disables opt-in safety checks. ## Required Notes - Existing global hard blocks remain active in every mode. - Careful mode is not a shell sandbox. - `LSTACK_CONFIRM_DESTRUCTIVE=1` or `lstack safety allow-once <hash>` can be used as an explicit override for opt-in careful/strict checks where appropriate. Global hard blocks still remain hard blocks.
From puukis/lstack, MIT.
More from puukis/lstack
- 0.471score
orchestrate
puukis
Intelligent task routing — evaluates complexity and asks the user whether to handle in the main session or dispatch to specialist sub-agents.
SkillCFreshness: CClaude Code +9 - 0.471score
interrogate
puukis
Socratic requirement clarification — one question at a time before any planning or code.
SkillCFreshness: CClaude Code +9 - 0.471score
blueprint
puukis
Generates a .blueprint.md spec file before any code is written.
SkillCFreshness: CClaude Code +9 - 0.469score
debug
puukis
Systematic root cause analysis — reproduce, isolate, hypothesize, verify before any fix
SkillCFreshness: CClaude Code +9 - 0.469score
docs
puukis
Writes accurate documentation by reading source first — never invents API signatures
SkillCFreshness: CClaude Code +9 - 0.469score
engineer
puukis
Writes production-ready code — reads patterns first, runs tests after, no gold-plating
SkillCFreshness: CClaude Code +9 - 0.469score
forget
puukis
Delete structured learnings or legacy observations from lstack memory.
SkillCFreshness: CClaude Code +9 - 0.469score
guard
puukis
Enable maximum safety workflow by combining careful safety mode with a freeze edit boundary.
SkillCFreshness: CClaude Code +9 - 0.469score
learn
puukis
Manually store a typed, trust-aware structured learning in lstack.
SkillCFreshness: CClaude Code +9 - 0.469score
parallel
puukis
Spawn isolated sub-agents across git worktrees for genuinely independent tasks
SkillCFreshness: CClaude Code +9 - 0.469score
planner
puukis
Structured task breakdown with clarifying questions before planning — no code written
SkillCFreshness: CClaude Code +9 - 0.469score
architect
puukis
System design before any code is written — invoked via /architect; produces ADRs and ARCHITECTURE.md
SkillCFreshness: CClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- May 24, 2026 (5 months ago)
- Last release
- May 18, 2026 (5 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/puukis/lstack
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from puukis
- 0.471score
blueprint
puukis
Generates a .blueprint.md spec file before any code is written.
SkillCFreshness: CClaude Code +9 - 0.471score
orchestrate
puukis
Intelligent task routing — evaluates complexity and asks the user whether to handle in the main session or dispatch to specialist sub-agents.
SkillCFreshness: CClaude Code +9 - 0.471score
interrogate
puukis
Socratic requirement clarification — one question at a time before any planning or code.
SkillCFreshness: CClaude Code +9 - 0.469score
debrief
puukis
End-of-session structured reflection — what worked, what broke, what to do differently.
SkillCFreshness: CClaude Code +9 - 0.469score
engineer
puukis
Writes production-ready code — reads patterns first, runs tests after, no gold-plating
SkillCFreshness: CClaude Code +9 - 0.469score
docs
puukis
Writes accurate documentation by reading source first — never invents API signatures
SkillCFreshness: CClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get careful running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
careful skill. Agent Mart. https://agentmart.dev/skills/puukis-careful (last updated 2026-10-24)