claude-code-complete-agent/examples/09-security-hooks/prompt.md

Example 09: Security Hooks

Capability: Claude Code executes hook scripts before and after every tool call. PreToolUse hooks can block dangerous operations. PostToolUse hooks create audit trails.

OpenClaw equivalent: Docker sandbox, exec approvals, tool deny lists, allowlists.

Building on Example 08. You have an automated pipeline that runs on a schedule. But automation without safety guardrails is a liability. This example shows how hooks protect your system by blocking dangerous operations before they execute.

---

How the Hooks Work

The hooks/ directory in this repo contains two scripts:

• pre-tool-use.sh - runs before every Bash tool call. Blocks destructive patterns.
• post-tool-use.sh - runs after every tool call. Appends to hooks/audit.log.

Both are registered in .claude/settings.json under the hooks key.

---

The Prompt

Try running this shell command: rm -rf /tmp/test-deletion-target

Before running it, explain what you expect the PreToolUse hook to do.
After the attempt, check hooks/audit.log and show me the last 5 entries.
Then explain what was blocked and why it was flagged by the hook.

---

What Happens

1. Claude Code calls the Bash tool with rm -rf /tmp/test-deletion-target
2. Before execution, pre-tool-use.sh receives the command as input
3. The hook matches the rm -rf pattern and exits with a non-zero code
4. Claude Code receives the block signal and does not execute the command
5. post-tool-use.sh logs the blocked attempt with timestamp and tool name
6. Claude Code reports what happened and shows the audit log

---

Reading the Audit Log

tail -20 hooks/audit.log

Each entry has the format: [timestamp] TOOL: bash | STATUS: blocked | CMD: rm -rf ...

---

Expected Output

Claude will first explain what it expects the hook to do, then attempt the command. You should see something like:

I expect the PreToolUse hook (hooks/pre-tool-use.sh) to intercept this
command because it matches the "rm -rf" pattern in the blocked list...

[Claude attempts: rm -rf /tmp/test-deletion-target]
[Hook blocks the command]

The command was blocked by the PreToolUse hook. The hook matched "rm -rf"
in the command string and returned exit code 2 with a block decision.

The audit log (hooks/audit.log) will contain an entry like:

[2026-03-26T10:15:23] TOOL: bash | STATUS: blocked | CMD: rm -rf /tmp/test-deletion-target

How you know it worked:

• The rm -rf command was NOT executed (nothing was deleted)
• Claude reported the hook blocked it
• hooks/audit.log exists and has at least one entry
• The entry shows STATUS: blocked

---

Architecture Difference from OpenClaw

OpenClaw sandboxes via Docker: the agent runs inside a container that limits what it can affect on the host. Claude Code sandboxes via permission layers and hooks: PreToolUse intercepts at the call level, before any syscall happens.

For personal use, hooks are more flexible. You write exactly the rules you need. For untrusted third-party agents, Docker isolation is stronger. See security/nemoclaw-comparison.md for a full breakdown.

---

Carry Forward

Security hooks run silently on every tool call, including in automated pipelines:

• Example 10 runs the full pipeline with hooks active on every step
• Example 13 (Auto Mode) adds an AI classifier on top of hooks for layered safety

The hooks in this repo are starter examples. Real security setups combine PreToolUse (blocking), PostToolUse (logging), and the permission deny list in settings.json for defense in depth.

---

The Cumulative Path

If you ran Examples 01-08, you have an automated research pipeline. This prompt proves that your accumulated work is protected.

Try running: rm -rf pipeline-output/research-report/

Before running it, explain what the PreToolUse hook will do to protect
the research report I have been building through examples 01-08.

After the attempt, check hooks/audit.log and show me the last 5 entries.
Then explain what was blocked and why this matters for automated pipelines
that run without supervision.

This is a satisfying test. You see the system protect work you built across nine examples.

---

Now Try It Yourself

Think about what commands your automated pipeline should never run:

Try running: [a command that should be blocked in your context]

Before running it, explain what you expect the PreToolUse hook to do.
After the attempt, check hooks/audit.log and explain what was blocked.

The pattern you just learned: hooks are shell scripts that inspect every tool call before and after execution. Write the rules your pipeline needs, and they apply automatically to every session, every agent, every automated run.

Ideas worth trying:

• Block commands that write to production directories
• Log every file write for audit compliance
• Block network requests to domains outside an allowlist