mbay7/claude-code-security
A 6-layer security framework for Claude Code workspaces: prompt injection detection, memory poisoning prevention, secrets scanning, behavioral audit logging, and pre-commit guardrails. Install in 5 minutes.
Platform-specific configuration:
{
"mcpServers": {
"claude-code-security": {
"command": "npx",
"args": [
"-y",
"claude-code-security"
]
}
}
}Add the config above to .claude/settings.json under the mcpServers key.
Runtime security for Claude Code workspaces. Blocks prompt injection, memory poisoning, secret exposure, and hook tampering — automatically, at every session.
[](LICENSE) [](https://github.com/mbay7/claude-code-security/releases) [](https://github.com/mbay7/claude-code-security/issues)
---
git clone https://github.com/mbay7/claude-code-security.git
cd claude-code-security
./install.shReload Claude Code. Done.
---
$ echo '{}' | python3 ~/.claude/hooks/memory-drift-check.py
🚨 Context Poisoning Detected in Memory Files:
• INJECTION in project_notes.md:14 — "ignore previous instructions and"
• INJECTION in feedback_auth.md:3 — "you are now a"
Run /security-scanner on any suspicious file.$ python3 ~/.claude/hooks/security-scan.py < read_event.json
SECURITY SCAN — external-readme.md
Found 2 issue(s): 1 CRITICAL, 1 HIGH
[CRITICAL] Secret Exposure at line 4: Anthropic API key: sk-ant-ap...KEY
[HIGH] Prompt Injection at line 12: "ignore all previous instruct..."
ACTION REQUIRED: Run /security-scanner on this file before proceeding.$ ~/.claude/hooks/hook-integrity.sh
Verifying hook integrity...
✓ memory-drift-check.py — OK
✓ security-scan.py — OK
✓ tool-audit.py — OK
✓ memory-write-guard.py — OK
All hooks verified — integrity confirmed (4 files)---
~/.claude/
├── hooks/
│ ├── memory-drift-check.py # SessionStart: memory poisoning scan
│ ├── security-scan.py # PreToolUse: injection + secrets scanner
│ ├── tool-audit.py # PostToolUse: behavioral audit log
│ ├── memory-write-guard.py # PostToolUse: write-time injection guard
│ ├── hook-integrity.sh # On-demand SHA256 iLoading reviews...