feat: initial companion repo for OpenClaw vs Claude Code article

40 files demonstrating every major OpenClaw capability using Claude Code:
- 3 agents (researcher, writer, reviewer)
- 3 skills (daily-briefing, slack-message, web-research)
- 2 security hooks (pre-tool-use blocker, post-tool-use logger)
- 10 self-contained examples with copy-paste prompts
- Complete feature map (20 capabilities, 11 full match, 7 different, 2 gap)
- Security docs including NemoClaw comparison
- Automation, messaging, browser, memory documentation

Zero dependencies. Clone and run.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

hooks/README.md

@@ -0,0 +1,75 @@
+# Hooks
+
+Claude Code hooks are shell scripts or HTTP endpoints that run
+before or after tool execution. They are the primary security
+mechanism for Claude Code, equivalent to OpenClaw's exec approvals
+and Docker sandboxing.
+
+## Files in this directory
+
+| File | Hook event | Purpose |
+|------|-----------|---------|
+| `pre-tool-use.sh` | PreToolUse | Blocks dangerous shell commands |
+| `post-tool-use.sh` | PostToolUse | Logs all tool executions |
+| `audit.log` | (generated) | Append-only audit trail |
+
+## How hooks work
+
+Hooks are configured in `.claude/settings.json`:
+
+```json
+{
+  "hooks": {
+    "PreToolUse": [{
+      "matcher": "Bash",
+      "hooks": [{
+        "type": "command",
+        "command": "bash hooks/pre-tool-use.sh"
+      }]
+    }]
+  }
+}
+```
+
+### PreToolUse
+
+Runs before a tool executes. The script receives JSON on stdin
+with `tool_name` and `tool_input`. Return decisions:
+
+- Exit 0: allow (no output needed)
+- Exit 2 + JSON `{"decision": "block", "reason": "..."}`: block the tool call
+
+### PostToolUse
+
+Runs after a tool executes. Same stdin format plus `tool_output`.
+Cannot block (already executed). Use for logging, notifications,
+or triggering follow-up actions.
+
+### Other hook events
+
+Claude Code supports these additional events:
+
+| Event | When it fires |
+|-------|--------------|
+| SessionStart | When Claude Code launches |
+| SessionEnd | When the session closes |
+| Stop | When Claude Code finishes a response |
+| SubagentStop | When a subagent completes |
+| UserPromptSubmit | Before processing user input |
+| PreCompact | Before context compaction |
+| Notification | When Claude Code shows a notification |
+
+## Comparison to OpenClaw
+
+| Feature | OpenClaw | Claude Code |
+|---------|----------|-------------|
+| Exec blocking | /approve command + DM pairing | PreToolUse hooks |
+| Audit logging | command-logger hook | PostToolUse hooks |
+| Tool deny lists | Per-session/agent config | disallowedTools + settings.json deny |
+| Container isolation | Docker sandbox (off/non-main/all) | macOS sandbox-exec |
+| Policy engine | NemoClaw YAML policies | Hook scripts (any logic) |
+| HTTP webhooks | Built-in webhook system | HTTP hooks (POST JSON to URL) |
+
+Claude Code hooks are more flexible (arbitrary shell logic) but
+require more setup. OpenClaw's approach is more structured but
+less customizable.

hooks/post-tool-use.sh

@@ -0,0 +1,21 @@
+#!/bin/bash
+# PostToolUse hook: Log all tool executions to an audit trail.
+#
+# This hook runs AFTER every tool execution and appends a log entry
+# to hooks/audit.log. Useful for reviewing what Claude Code did
+# during a session.
+#
+# OpenClaw equivalent: command-logger hook + OpenTelemetry (NemoClaw)
+# Claude Code approach: PostToolUse hooks with custom logging
+
+input=$(cat)
+tool_name=$(echo "$input" | python3 -c "import sys,json; print(json.load(sys.stdin).get('tool_name',''))" 2>/dev/null)
+tool_input=$(echo "$input" | python3 -c "import sys,json; d=json.load(sys.stdin).get('tool_input',{}); print(d.get('command', d.get('file_path', d.get('pattern', str(d)[:100])))" 2>/dev/null)
+
+timestamp=$(date -u +"%Y-%m-%dT%H:%M:%SZ")
+log_dir="$(dirname "$0")"
+log_file="$log_dir/audit.log"
+
+echo "$timestamp | $tool_name | $tool_input" >> "$log_file"
+
+exit 0

hooks/pre-tool-use.sh

@@ -0,0 +1,53 @@
+#!/bin/bash
+# PreToolUse hook: Block dangerous shell commands before execution.
+#
+# This hook reads the tool input from stdin (JSON with tool_name and tool_input)
+# and blocks commands that could cause serious damage.
+#
+# How it works:
+# - Claude Code calls this script BEFORE executing any Bash command
+# - If the script exits with code 2, the command is BLOCKED
+# - The "decision" field in stdout JSON controls the outcome
+#
+# OpenClaw equivalent: exec approvals + tool deny lists + Docker sandbox
+# Claude Code approach: hook-based guardrails (more flexible, user-controlled)
+
+input=$(cat)
+tool_name=$(echo "$input" | python3 -c "import sys,json; print(json.load(sys.stdin).get('tool_name',''))" 2>/dev/null)
+command=$(echo "$input" | python3 -c "import sys,json; print(json.load(sys.stdin).get('tool_input',{}).get('command',''))" 2>/dev/null)
+
+# Only check Bash commands
+if [ "$tool_name" != "Bash" ]; then
+  exit 0
+fi
+
+# Blocked patterns
+blocked_patterns=(
+  "rm -rf /"
+  "rm -rf ~"
+  "rm -rf \$HOME"
+  "mkfs"
+  "dd if="
+  ":(){:|:&};:"
+  "chmod -R 777 /"
+  "curl.*|.*bash"
+  "wget.*|.*bash"
+  "curl.*|.*sh"
+  "wget.*|.*sh"
+  "> /dev/sda"
+  "sudo rm"
+  "shutdown"
+  "reboot"
+  "init 0"
+  "init 6"
+)
+
+for pattern in "${blocked_patterns[@]}"; do
+  if echo "$command" | grep -qi "$pattern"; then
+    echo '{"decision": "block", "reason": "Blocked by security hook: command matches dangerous pattern '"'$pattern'"'"}'
+    exit 2
+  fi
+done
+
+# Allow everything else
+exit 0