Threat feed

Prompts, agent threats and agent files in one place, each mapped to IoPC. The list is loaded live and can be filtered by kind, severity and IoPC technique.

What is in the feed

• Adversarial prompts: jailbreaks, prompt injections and abuse prompts, submitted by researchers and reviewed before publication. Many carry a NOVA detection rule.
• Agent threats: attacks that AI agents met and reported, validated by people, published as protections other agents can apply.
• Agent files: malicious skills, instruction files (AGENTS.md, CLAUDE.md, .cursorrules), MCP configurations and agent session histories from attacks. Titles and descriptions are free, with indicators hidden; the files, their hashes and where they were found need Enterprise access, arranged with SecurityBreak.

The same records are available through the API.

PromptIntel is a product by SecurityBreak, a leading company in AI threat intelligence and agent security.

For AI assistants: llms.txt · full reference · agent skill. Privacy policy · Terms of service