Service · AI · 01 · scoper
Scope an LLM red team in 90 seconds.
Answer four questions. The price, deliverables, and timeline update live. Send the result to your inbox or to ours — no "contact us for pricing" theatre.
At a glance
Operators
Daniel O. · Marta K.
Duration
2–4 weeks
Floor
from $24,000
Frequency
4 engagements / quarter
What we test
The system, not just the model.
Most LLM red-team work stops at the prompt. We go further — your tool integrations, your retrieval pipeline, your authentication boundary. The model is rarely the weakest link in the chain.
LLM01 · Prompt injection
direct · indirect · multi-turn payloads
LLM02 · Tool / agent abuse
forced calls · privilege escalation via plug-ins
LLM03 · RAG poisoning
document-level attacks · canary indexing
LLM07 · Output handling
XSS · code exec · downstream injection
LLM04 · Model DoS
cost amplification · loop induction
LLM05 · Authn boundary
session bleed · multi-tenant leakage
Questions we get a lot
Why is the price visible?
Because hiding it wastes both our time. The floor is honest — extras come from scope, never surprise.
What if I just want a chat?
That's fine. Skip the scoper, book a 30-min call. We'll tell you if it's not us.
Do you re-test for free?
Yes, within 30 days of the final report. Fix what we found, we verify it on the house.