Developer guardrails

Setup Git Guardrails

Sets up a PreToolUse hook that intercepts and blocks dangerous git commands before Claude executes them.

What Gets Blocked

• git push (all variants including --force)
• git reset --hard
• git clean -f / git clean -fd
• git branch -D
• git checkout . / git restore .

When blocked, Claude sees a message telling it that it does not have authority to access these commands.

Steps

1. Ask scope

Ask the user: install for this project only (.claude/settings.json) or all projects (~/.claude/settings.json)?

2. Copy the hook script

The bundled script is at: scripts/block-dangerous-git.sh

Copy it to the target location based on scope:

• Project: .claude/hooks/block-dangerous-git.sh
• Global: ~/.claude/hooks/block-dangerous-git.sh

Make it executable with chmod +x.

3. Add hook to settings

Add to the appropriate settings file:

Project (.claude/settings.json):

{
  "hooks": {
    "PreToolUse": [
      {
        "matcher": "Bash",
        "hooks": [
          {
            "type": "command",
            "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/block-dangerous-git.sh"
          }
        ]
      }
    ]
  }
}

Global (~/.claude/settings.json):

{
  "hooks": {
    "PreToolUse": [
      {
        "matcher": "Bash",
        "hooks": [
          {
            "type": "command",
            "command": "~/.claude/hooks/block-dangerous-git.sh"
          }
        ]
      }
    ]
  }
}

If the settings file already exists, merge the hook into existing hooks.PreToolUse array — don't overwrite other settings.

4. Ask about customization

Ask if user wants to add or remove any patterns from the blocked list. Edit the copied script accordingly.

5. Verify

Run a quick test:

echo '{"tool_input":{"command":"git push origin main"}}' | <path-to-script>

Should exit with code 2 and print a BLOCKED message to stderr.

---

setup-pre-commit

Setup Pre-Commit Hooks

What This Sets Up

• Husky pre-commit hook
• lint-staged running Prettier on all staged files
• Prettier config (if missing)
• typecheck and test scripts in the pre-commit hook

Steps

1. Detect package manager

Check for package-lock.json (npm), pnpm-lock.yaml (pnpm), yarn.lock (yarn), bun.lockb (bun). Use whichever is present. Default to npm if unclear.

2. Install dependencies

Install as devDependencies:

husky lint-staged prettier

3. Initialize Husky

npx husky init

This creates .husky/ dir and adds prepare: "husky" to package.json.

4. Create .husky/pre-commit

Write this file (no shebang needed for Husky v9+):

npx lint-staged
npm run typecheck
npm run test

Adapt: Replace npm with detected package manager. If repo has no typecheck or test script in package.json, omit those lines and tell the user.

5. Create .lintstagedrc

{
  "*": "prettier --ignore-unknown --write"
}

6. Create .prettierrc (if missing)

Only create if no Prettier config exists. Use these defaults:

{
  "useTabs": false,
  "tabWidth": 2,
  "printWidth": 80,
  "singleQuote": false,
  "trailingComma": "es5",
  "semi": true,
  "arrowParens": "always"
}

7. Verify

• .husky/pre-commit exists and is executable
• .lintstagedrc exists
• prepare script in package.json is "husky"
• prettier config exists
• Run npx lint-staged to verify it works

8. Commit

Stage all changed/created files and commit with message: Add pre-commit hooks (husky + lint-staged + prettier)

This will run through the new pre-commit hooks — a good smoke test that everything works.

Notes

• Husky v9+ doesn't need shebangs in hook files
• prettier --ignore-unknown skips files Prettier can't parse (images, etc.)
• The pre-commit runs lint-staged first (fast, staged-only), then full typecheck and tests

---

Merged from: safety-guardrails

Step 0: Gather project context

_ROOT=$(git rev-parse --show-toplevel 2>/dev/null || pwd)
_SLUG=$(basename "$_ROOT")
_BRANCH=$(git branch --show-current 2>/dev/null || echo "main")
_MEM="$_ROOT/memory"
mkdir -p "$_MEM" "$_MEM/sessions" "$_MEM/checkpoints" "$_MEM/retros" "$_MEM/reviews" "$_MEM/specs"
echo "=== Context: $_SLUG / $_BRANCH ==="
[ -f "$_MEM/context.md" ]      && echo "--- last context ---"     && tail -30 "$_MEM/context.md"
[ -f "$_MEM/learnings.jsonl" ] && echo "--- recent learnings ---" && tail -5  "$_MEM/learnings.jsonl"
[ -f "$_MEM/timeline.jsonl" ]  && echo "--- recent timeline ---"  && tail -5  "$_MEM/timeline.jsonl"

Memory dir (memory/): replaces gbrain. grep -r "X" memory/ ≡ gbrain search X · echo '...' >> memory/timeline.jsonl ≡ gbrain store

---

Production Safety Guard (SKC)

Safety Guard — Prevent Destructive Operations

When to Use

• When working on production systems
• When agents are running autonomously (full-auto mode)
• When you want to restrict edits to a specific directory
• During sensitive operations (migrations, deploys, data changes)

How It Works

Three modes of protection:

Mode 1: Careful Mode

Intercepts destructive commands before execution and warns:

Watched patterns:
- rm -rf (especially /, ~, or project root)
- git push --force
- git reset --hard
- git checkout . (discard all changes)
- DROP TABLE / DROP DATABASE
- docker system prune
- kubectl delete
- chmod 777
- sudo rm
- npm publish (accidental publishes)
- Any command with --no-verify

When detected: shows what the command does, asks for confirmation, suggests safer alternative.

Mode 2: Freeze Mode

Locks file edits to a specific directory tree:

/safety-guard freeze src/components/

Any Write/Edit outside src/components/ is blocked with an explanation. Useful when you want an agent to focus on one area without touching unrelated code.

Mode 3: Guard Mode (Careful + Freeze combined)

Both protections active. Maximum safety for autonomous agents.

/safety-guard guard --dir src/api/ --allow-read-all

Agents can read anything but only write to src/api/. Destructive commands are blocked everywhere.

Unlock

/safety-guard off

Implementation

Uses PreToolUse hooks to intercept Bash, Write, Edit, and MultiEdit tool calls. Checks the command/path against the active rules before allowing execution.

Integration

• Enable by default for codex -a never sessions
• Pair with observability risk scoring in SKC 2.0
• Logs all blocked actions to ~/.claude/safety-guard.log