Anti-patterns · #7
Cascading Security Failures
- Symptom
- A poisoned doc in a shared index contaminates every downstream consumer. Prompt injection propagates via handoffs.
- Fix
- Treat every agent-to-agent message as a trust boundary. Partition knowledge bases by trust level.
- Origin
- Introduced at L3
- Cross-link
- The architecture-level failure the Security threat model treats end to end: every OWASP LLM and Agentic threat that propagates across a trust boundary traces back here. Knowledge-base counterpart: Part IX §7.
Keep exploring
Now the right way.
PatternsControllerContinuously monitors the behaviour of the agent and transparently aligns it with ethical principles and compliance rules. Realised at scale as a governance agent (watching for policy violations) or security agent (flagging anomalous behaviour) supervising other agents, escalating to a human only on a trip.Read Controller SecuritySecurity Threat ModelThe OWASP LLM and Agentic Threats catalogs, mapped to the defenses this project documents.See the threat model PatternsAnti-PatternsThe failure modes to design against.Back to anti-patterns