Search

Run scoped, authorized Sn1per pentest, recon and attack surface scans from AI agents; read findings.

Drop-in auth gateway for self-hosted MCP servers, API-key auth, per-tool allowlists, rate limiting & audit logs, with zero upstream changes. Single Go binary.

Agentic recon CLI: RAG-grounded LLM drives 6 live tools (VisorGraph, aimap, BARE, nuclei, Menlo-hunt, OSV-scan) with every probe sandboxed in gVisor

On-chain Solana token safety — screen for rugs/honeypots and execute MEV-protected swaps.

Noisegate: a differential privacy gateway that lets an untrusted LLM agent query sensitive data over MCP (Model Context Protocol), with a formal guarantee no individual's record can leak even if the agent is adversarial - enforcement lives in trusted code below the model, validated by a runnable attack gallery.

Local-first PHI detection for source code, LLM prompts, logs, and analytics

Facilitates vulnerability scanning on various LLMs using Garak through a lightweight MCP server.

CLI and MCP servers for Cutline — authenticate, then run constraint-aware MCP servers in Cursor or any MCP client.

Verifiable cryptographic identity and delegation for AI agents across MCP and A2A

Local LLM prompt injection firewall

A local, deterministic firewall for AI agent tool calls: Claude Code hooks, an MCP proxy and a Python API, with replayable policies for CI

Kernel-to-MCP flight recorder for AI agents on macOS: attributes every exec, file access and socket to the agent and MCP tool call that caused it

🐍 ☁️ - MCP server for interacting with the CyberChef server API which will allow an MCP client to utilise the CyberChef operations.

MCP Permission Auditor — scan, enumerate, and risk-score all locally configured MCP servers

Scans any wallet for dangerous ERC-20/721/1155 token approvals and infinite allowances, scoring drainer exposure and emitting revoke transactions.

A safety floor your AI agents can't get under: one policy for Claude Code, Codex, Gemini CLI, Cursor, Windsurf and MCP, judged by what each call will run, with a tamper-evident, signed ledger.

Refactoring Assistant MCP - Extract functions, convert callbacks to async/await, simplify conditionals, and apply design patterns. Works with Claude Code, Cursor, Windsurf, Roo Code, and any MCP-compatible editor.

Skill router and prompt-injection firewall for AI coding agents: give Claude Code, Codex and Cursor the right skills, MCP servers and tools for each task.

WHOIS/RDAP, DNS, SSL, live subdomains with IPs, and SPF/DMARC/DKIM for any domain.

MCP server for remote SSH operations -- persistent sessions, structured command execution, SFTP file transfer, and port forwarding for AI agents.