
pentagi
AgentSelf-hosted multi-agent system that plans and runs penetration tests in a sandbox
- Open source
- Open-source agent
- 34.3K GitHub stars
- Security

Self-hosted multi-agent system that plans and runs penetration tests in a sandbox
Sign agent actions into a tamper-evident log and verify any agent's log from an MCP client
Remote MCP server for Curve Compliance's healthcare marketing and consent reports
Not checked yetLet an AI agent register and pay for its own verifiable DID-LLM identity
A safer MCP for Dokploy: agents deploy and fix apps without seeing keys or secrets
Not checked yetAsk Claude about on-site incidents, monthly counts and trends in SISA
Bulk domain lookup: tech stack, email provider, DNS, WHOIS age, SSL expiry, AI bot rules
Local, reversible masking of client names and numbers before text goes to an AI
Runs locallyReal-time alerts for deepfake voices, impersonation and social engineering on live calls
Not checked yet
Open-source control plane that firewalls MCP tool calls and caps agent model spend
Flags fake engagement and prompt injection in AI-agent job board listings
Runs locallyPseudonymise or redact German court rulings and filings locally before AI sees them
Runs locallyList AugmentEV's agents; API key required and accounts are by request
Not checked yet
Allow-listed SQL reads and allow-or-deny checks on writes for agents
Not checked yetAsk business questions across connected enterprise systems with governed definitions
Not checked yetDarknet, Telegram and ransomware threat intelligence queried from AI assistants
Not checked yetGoverned gateway that puts ERP, CRM and HR apps behind Claude, ChatGPT and Copilot
Not checked yetStatic security scanner for MCP servers, agent skills and npm or PyPI packages
Runs locallyTriage brand impersonators across domains, trademarks, apps and ads from an agent
Not checked yet
Bulk domain age, registrar, expiry, DNS, email setup, Wayback history and SSL checks
Not checked yetOffline security checks for code, dependencies and AI agent setup with six read-only tools
Runs locallyRun ad scans, write detection rules and manage alerts across 108 tools
Not checked yetHosted sign-in, OIDC and API keys so agents and apps can confirm who a user is
Not checked yetRegistry, scoped permissions and one-step recall for AI agents in health systems
Not checked yet
Scan prompts, tool definitions and model output for injection, and guard tool calls
Runs locally
Pay-per-call read-only crypto data and security checks, settled in USDC on Base via x402
Not checked yetPassive-by-default OSINT lookups for AI agents: DNS, WHOIS, IP, breaches, Shodan and more
Runs locallyAsk Copilot or Claude to audit and operate client Microsoft 365 tenants as an MSP
Not checked yetSeven pay-per-call x402 tools for drug safety, token risk scans and Base escrow checks
Not checked yetScan text for leaked keys, emails and your own private terms before sharing
Not checked yet