LLM Guardrail Breaker
An adversarial testing framework for AI agents, MCP servers, and RAG workflows covering injection, tool poisoning, SSRF, and data exfiltration.
- Role
- Solo build — research, engineering, documentation, release
- Status
- Planned · frontier build
- Year
- 2026
- Position
- 13 of 13
01 — Context
The problem worth solving.
Agent security fails at the boundary between model output, tools, data retrieval, and user privileges.
02 — Approach
How it was built.
- 01
Automate indirect prompt injection and cross-tenant retrieval tests.
- 02
Test SSRF through agent tools and malicious MCP schemas.
- 03
Produce risk reports plus scoped tool wrappers and deterministic controls.
03 — Architecture
The shape of the system.
Agent, MCP server, retrieval layer, tools, and privilege context.
Injection, poisoning, SSRF, and data-exfiltration scenarios.
Input/output firewalls, sandboxing, and least-privilege tool wrappers.
04 — Evidence
Receipts, not adjectives.
05 — Hiring signal
“Treats agent security as an end-to-end systems problem instead of a prompt-only problem.”
06 — What's next
Still being built.
Define the attack corpus and implement the first tool-poisoning harness.