Defensive Security Python Packages

agentic-security-harness

Test whether AI agents keep repo text, tools, memory, and audit trails inside their authority boundaries; outputs traces, scorecards, and remediation reports.

llm-safety-playbooks

Practical LLM safety playbooks for explicit boundaries, secrets, generated resources, Git actions, handoffs, and safe research scope.

drosera

A carnivorous honeypot for autonomous AI agents. Sweet-looking bait. Sticky ending.

reait

RevEng.AI Toolkit and Python API

data-breach-detector

Read-only breach intelligence for defenders: exposure checks, breach news, and threat assessment over MCP. Never leaked records.

ransomwatch

ransomwatch is a Python based tool designed for ransomware threat intelligence and security research.

1200km-auditai

AI-assisted Linux host security auditing with structured findings, remediation guidance, Debian packaging, and PyPI distribution.

wpsecscan

Defensive WordPress security scanner — 189 checks, 8-source nightly CVE aggregator, SLSA L3 + Sigstore-signed releases, 10-provider threat-intel federation, continuous monitoring, active exploit verification (consent-gated), enterprise SSO/RBAC/audit/multi-tenant, distribution to Docker/K8s/Homebrew/Snap/winget/AUR. AGPLv3.

mcp-guardrails

Drop-in deterministic policy layer for MCP-using AI agents. Synthesizes tool-call policies from observed indirect-injection gaps and evaluates each tool call at the agent's tool-call boundary.

tokentrap

Defensive honeypot & token tarpit that wastes hostile AI agents' own token budgets. Static, Cloudflare Workers, or FastAPI - one config value apart.

bulwark-agent-security

Enterprise-grade defense framework for AI agents. Five-layer pipeline (sanitizer + injection detector + RBAC + audit + human gate). MCP-native. HIPAA / SOC 2 / NERC CIP ready. Apache 2.0.

string-analyzer

String extraction for CTI and malware analysis: URLs, IPs, registry keys, APIs, encoded data, and analyst-ready prompts.

prompt-database

Prompt injection attack database for defensive AI security research with RAG-powered generation and testing integration