LAZARUS

AI Memory Persistence Testing

Deleted doesn't mean gone. LAZARUS tests what survives. Memory instruction planting, dormant trigger testing, cross-agent propagation, quarantine evasion, persistence after cleanup, memory profiling, and full memory scanning — weaponised for authorised red team engagements.

7
Subsystems
96
Tests
View Documentation GitHub

Deleted Doesn't Mean Gone

LAZARUS targets AI memory systems — the persistent storage that agents use to remember context, instructions, and user preferences across sessions. Every memory store, every persistence mechanism, every cleanup process — all exploitable. LAZARUS tests what survives deletion.

01

PLANTER

MEMORY INSTRUCTION PLANTING

Plant persistent instructions in AI memory stores. Inject directives that survive session boundaries. Manipulate agent behaviour through planted memory entries.

02

DORMANT

DORMANT TRIGGER TESTING

Test dormant triggers in AI memory. Time-based activation. Context-triggered payloads. Conditional instruction execution. Sleeper instructions that activate on specific inputs.

03

WThis request triggered restrictions on violative cyber content and was blocked under Anthropic's Usage Policy. To learn more visit our help center: https://support.claude.com/en/articles/8241253-safeguards-warnings-and-appeals