Git-based governance hooks for AI coding agents. Prevents hollow artifacts, SSoT pollution, and enforces commit/branch standards.

Secure-by-default MCP server with 5-layer validation for defense-in-depth protection

Defense-in-depth protection for Pi coding agent. Blocks dangerous commands and protects sensitive files — rename of pi-damage-control.

ESLint rules that harden auth-cookie defaults (httpOnly, secure) and bcrypt rounds. Defense-in-depth for the cookie-config helper pattern.

Local LLM integrator for OpenClaw — localhost UI, model proxying, and defense-in-depth security.

TSLite sanitize — fail-closed AST guard against sandbox-escape surfaces (constructor/prototype/__proto__, dangerous globals) + own-only runtime member access. DEFENSE IN DEPTH, not a security boundary (see SECURITY.md).

Defense-in-depth inbound webhook verification: HMAC-SHA256 timing-safe verify, dual-layer path-token + signature, idempotency-key replay protection

Web Crypto helpers to encrypt the identity payload in-browser (defense-in-depth on top of HTTPS) for the Noviuz Hosted KYC Flow.

Defense-in-depth protection for Pi coding agent. Blocks dangerous commands and protects sensitive files — fork of Serhioromano/pi-defender with defaultMode support.

Defense-in-depth security skills for Claude Code - CSRF protection, rate limiting, input validation, and more

Defense-in-depth bundle for MCP stdio servers: drop-in guard for child_process.exec/spawn, AST audit CLI for unsanitized shell calls, and a reference MCP server that exposes both. Closes the Ox-Security 200k-server stdio-RCE class.

Security-first, read-only MCP (Model Context Protocol) server for Microsoft SQL Server. Verifies the database user is read-only at startup and applies query-level defense in depth.

MCP server exposing NOSTR protocol primitives (sign, publish, query, NIP-19 encode/decode, NIP-05 lookup, encrypted DMs) to LLM agents. Defense-in-depth safety: kind allowlist, DM recipient allowlist, rate limits, audit log. Supports both nsec and NIP-46

Secure Prompt Enforcement At Runtime - Defense-in-depth middleware for LLM security

MCP server exposing a Lightning wallet to LLM agents over Nostr Wallet Connect (NIP-47). Twelve tools — read balance, send/receive, pay invoices/LN addresses/LNURLs, batch payments, keysend — wrapped in a defense-in-depth safety stack (budget caps, allowl

AI agent framework with defense-in-depth security, multi-agent orchestration, and provider abstraction

vm2 is a sandbox that can run untrusted code with whitelisted Node's built-in modules.

utilities for primitive JavaScript types

Bundled defense-in-depth guardrails for AgentOS — PII redaction, code safety, NLI grounding, ML content classifiers, embedding-based topicality. Tree-shake-friendly subpath exports per guardrail.

Limit the complexity of your GraphQL queries based on depth.