Start here
I need to…
Pick the job in front of you. Each playbook gives the patterns practitioners agree on, an open-source stack from the radar, the questions to ask vendors, a checklist you can tick off, and the news that changes it. Checklists are saved in this browser.
- Red-team an LLM application AppSec, pentesters, AI platform teams You have a chatbot, copilot or agent going to production and need evidence it resists prompt injection, data leakage and tool misuse. 9 steps 30 recent stories
- Secure AI coding tools in your organisation Security engineering, IT, developer platform teams Developers run Claude Code, Cursor, Copilot, Codex and similar agents with access to source, terminals, cloud credentials and MCP servers. 8 steps 25 recent stories
- Lock down MCP servers and clients Security engineering, platform teams, developers building agents MCP lets agents call tools from many publishers. 8 steps 30 recent stories
- Secure RAG and vector stores AppSec, AI platform and data teams Retrieval-augmented apps pull documents, tickets and web pages into the model's context. 8 steps 18 recent stories
- Give AI agents identity and least privilege IAM, security architecture, platform teams Agents usually run with a shared service account or inherit everything the invoking user can do. 8 steps 27 recent stories
- Get visibility into Shadow AI Security operations, GRC, IT Employees use chat assistants, browser extensions, AI features inside SaaS apps and personal accounts for work data. 7 steps 9 recent stories
- Respond to a compromised AI agent Incident response, SOC, platform owners An agent or AI coding assistant followed injected instructions, ran a poisoned MCP tool or leaked data through a tool call. 9 steps 30 recent stories
- Evaluate an AI SOC or agentic triage tool SOC leads, detection engineering, CISOs Vendors pitch AI analysts that triage alerts and run investigations. 7 steps 2 recent stories
- Prepare for AI-powered attackers Vulnerability management, SOC, red teams Autonomous offensive agents now find and validate web vulnerabilities, and threat actors have used open-source agent frameworks in automated intrusions against many organisations at once. 5 steps 10 recent stories
Playbooks are written by the editor from practitioner discussions and public guidance, and reviewed periodically; tools and news on each page update automatically.