Code Firewall Mcp
A structural pattern matcher that sits between Claude and code execution tools like rlm_exec. It parses code into syntax trees via tree-sitter, normalizes away identifiers and literals to expose the underlying structure, embeds the result through Ollama, then checks similarity against a ChromaDB blacklist of known-bad patterns. The idea is that os.system("rm -rf /") and os.system("ls") have identical structure, so you can catch dangerous patterns regardless of specific arguments. You'd use this when giving Claude access to shell or exec tools and want a secondary filter beyond the model's own judgment. Ships with setup tools for Ollama, blacklist management via firewall_blacklist, and both file and string checking methods. Threshold defaults to 0.85 similarity but you can tune it.
- tools
- 11
- transport
- STDIO
- env vars
- 4
- last updated
- Jan 19, 2026
Tools
11 toolsPublic tool metadata for what this MCP can expose to an agent.
firewall_system_checkCheck system requirements — verify macOS, Apple Silicon, RAM
Check system requirements — verify macOS, Apple Silicon, RAM
No parameter schema in public metadata yet.
firewall_setup_ollamaInstall via Homebrew — managed service, auto-updates
Install via Homebrew — managed service, auto-updates
No parameter schema in public metadata yet.
firewall_setup_ollama_directInstall via direct download — no sudo, fully headless
Install via direct download — no sudo, fully headless
No parameter schema in public metadata yet.
firewall_ollama_statusCheck Ollama availability — verify embeddings are ready
Check Ollama availability — verify embeddings are ready
No parameter schema in public metadata yet.
firewall_checkCheck if a code file is safe to pass to execution tools.
Check if a code file is safe to pass to execution tools.
No parameter schema in public metadata yet.
firewall_check_codeCheck code string directly (no file required).
Check code string directly (no file required).
No parameter schema in public metadata yet.
firewall_blacklistAdd a dangerous pattern to the blacklist.
Add a dangerous pattern to the blacklist.
No parameter schema in public metadata yet.
firewall_record_deltaRecord near-miss variants to sharpen the classifier.
Record near-miss variants to sharpen the classifier.
No parameter schema in public metadata yet.
firewall_list_patternsList patterns in the blacklist or delta collection.
List patterns in the blacklist or delta collection.
No parameter schema in public metadata yet.
firewall_remove_patternRemove a pattern from blacklist or deltas.
Remove a pattern from blacklist or deltas.
No parameter schema in public metadata yet.
firewall_statusGet firewall status and statistics.
Get firewall status and statistics.
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageConfiguration
FIREWALL_DATA_DIRDirectory for storing ChromaDB data
OLLAMA_URLURL for Ollama server (default: http://localhost:11434)
EMBEDDING_MODELOllama embedding model (default: nomic-embed-text)
SIMILARITY_THRESHOLDBlock threshold 0-1 (default: 0.85)
Readme
Details
- Repository
- github.com/egoughnour/code-firewall-mcp
- Category
- Security & Pentesting
- Package
code-firewall-mcp- Transport
- STDIO
- Registry
- MCP registry, active
More from egoughnour
Related Security & Pentesting MCP servers
View all
Kloudle Cloud Security Scannerio.github.kloudle/cloud-security-scannerAWS cloud security scanners for AI agents — S3, IAM, EC2, EKS, RDS, CloudTrail, CloudWatch Logs0stars
Agent Securitymdfifty50-boop/agent-security-mcpSecurity scanning and threat detection for AI agents0stars
Panda Web Security Toolspandadad0214/panda-web-security-toolsSite/email/agent safety and monitors: SPF/DKIM, SEO, change + RSS monitors, hidden-AI, WHOIS (7).0stars
Security Orchestrarobotfleet-hq/security-orchestraMulti-agent MCP platform for data centers and critical power.0stars
GrindFree Agent Securityronin93/grindfree-agent-securityPay-per-call agent security (x402): injection, tools, shell/SQL, memory, leaks, docs, safe web reads0stars
Vulnicheckandrasfe/vulnicheckHTTP MCP Server for comprehensive Python vulnerability scanning and security analysis.11stars