Code Firewall Mcp
by egoughnourcommunityv0.7.0
Code Firewall MCP is a structural similarity-based code security filter that blocks dangerous code patterns before execution by comparing code structure against a blacklist of known-bad patterns.
From the registry: Structural similarity-based code filter. Stops malicious code pattern reaching execution tools.
Install & configure
Prerequisites
- requiredPython 3.10+Required Python version (< 3.14 due to onnxruntime).
- required
- requiredOllamaInstall and run Ollama with the embedding model: `ollama pull nomic-embed-text`. The MCP server has setup tools to install Ollama on macOS.
https://ollama.ai$ ollama pull nomic-embed-text
Install in your client
Paste into the AI you're chatting with — it will figure out the rest
Please install the `code-firewall-mcp` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Python 3.10+** — Required Python version (< 3.14 due to onnxruntime).
- **uv (for uvx)** — Install uv to run via uvx. (https://docs.astral.sh/uv/)
- **Ollama** — Install and run Ollama with the embedding model: `ollama pull nomic-embed-text`. The MCP server has setup tools to install Ollama on macOS. Run: `ollama pull nomic-embed-text` (https://ollama.ai)
Canonical MCP server config (stdio transport):
- command: `uvx`
- args: ["code-firewall-mcp"]
- optional environment variables:
- `FIREWALL_DATA_DIR`: Data storage directory (default /tmp/code-firewall). (example: `~/.code-firewall`)
- `OLLAMA_URL`: Ollama server URL (default http://localhost:11434). (example: `http://localhost:11434`)
- `EMBEDDING_MODEL`: Ollama embedding model (default nomic-embed-text). (example: `nomic-embed-text`)
- `SIMILARITY_THRESHOLD`: Block threshold 0-1 (default 0.85). (example: `0.85`)
- `NEAR_MISS_THRESHOLD`: Near-miss recording threshold (default 0.70). (example: `0.70`)
Note: Uses ChromaDB locally for vector storage. Built around macOS Apple Silicon.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.Note: Uses ChromaDB locally for vector storage. Built around macOS Apple Silicon.
Tools (11)
firewall_system_check
Check system requirements — verify macOS, Apple Silicon, RAM
firewall_setup_ollama
Install via Homebrew — managed service, auto-updates
firewall_setup_ollama_direct
Install via direct download — no sudo, fully headless
firewall_ollama_status
Check Ollama availability — verify embeddings are ready
firewall_check
Check if a code file is safe to execute
firewall_check_code
Check code string directly (no file required)
firewall_blacklist
Add a dangerous pattern to the blacklist
firewall_record_delta
Record near-miss variants for classifier sharpening
firewall_list_patterns
List patterns in blacklist or delta collection
firewall_remove_pattern
Remove a pattern from blacklist or deltas
firewall_status
Get firewall status and statistics
Environment variables
FIREWALL_DATA_DIRrequiredData storage directoryOLLAMA_URLrequiredOllama server URLEMBEDDING_MODELOllama embedding modelSIMILARITY_THRESHOLDBlock threshold (0-1)NEAR_MISS_THRESHOLDNear-miss recording thresholdPrerequisites
- • Python 3.10+
Packages
code-firewall-mcp
pypiv0.7.0transport: stdio
https://github.com/egoughnour/code-firewall-mcp/releases/download/v0.7.0/code-firewall-mcp.mcpb
mcpbtransport: stdio
Details
- Language
- Python
- License
- MIT
- Released
- Jan 19, 2026
- Last commit
- Jan 19, 2026
- Registry name
- io.github.egoughnour/code-firewall-mcp