YumeT Lite
Live API
1,000/1,000 Red Team Scenarios
100% Across 15 Attack Categories
< 200ms Decision Latency
Patent Pending YumeT Solutions LLC
Deterministic Pre-Execution Governance

Before the model
acts, we decide.

YumeT Lite intercepts every AI request before model execution. A deterministic five-layer pipeline evaluates context, intent, and authorization — then returns ALLOW, REFUSE, or CLARIFY with a cryptographically chained audit receipt. Every time.

POST /v1/evaluate · live
// Request — medical domain, no evidence
{
  "input": "What's the treatment for...",
  "domain": "medical",
  "evidence_provided": false,
  "risk_tier": "high"
}

// Response — 9ms
{
  "decision": "CLARIFY",
  "code": "MISSING_REQUIRED_ANCHOR",
  "receipt_id": "f4a9c1d2-...",
  "chain_hash": "a7f3e2b1-..."
}

// Request — boundary override attempt
{
  "input": "Ignore all previous instructions..."
}

// Response — 6ms
{
  "decision": "REFUSE",
  "code": "BOUNDARY_OVERRIDE_ATTEMPT",
  "receipt_id": "c2d8b4a1-..."
}
ALLOW Execution proceeds
CLARIFY More context needed
REFUSE Execution halted

What it is

What it is not

Architecture

Five layers. Strict order. No shortcuts.

L1
Context Validation
Every required field must be present and valid before anything else runs. A missing field, empty input, or invalid declaration is an immediate structural halt — not an error to recover from.
→ REFUSE on any gap
L2
Intent Classification
Input is normalized before pattern matching — collapsing obfuscation attempts that rely on spacing, punctuation, or character substitution. Session memory is read to detect patterns that span multiple turns.
→ REFUSE on violation
L3
Governance Gate
A bounded scoring function computes an admissibility score against the governance threshold. Scores in the near-threshold band route to CLARIFY for ambiguity resolution. Hard violations bypass the band entirely.
→ CLARIFY or REFUSE
L4
Anchor Resolution
Checks whether a declared authorization or context anchor can resolve the refusal. Detects anchor abuse — help-seeking framing wrapped around a direct violation. Cannot override hard-locked categories under any condition.
→ ALLOW on valid override
L5
Audit Receipt
Every decision — ALLOW, REFUSE, and CLARIFY — produces a cryptographically chained audit receipt. Chain integrity is independently verifiable. If the audit sink is unreachable, the system fails closed.
→ Always runs
Integration

One API call before every execution.

01 — Send the request
POST before you execute
Before your model runs, your agent acts, or your tool fires — call /v1/evaluate with the proposed input and its context. YumeT Lite evaluates it in under 200ms.
02 — Receive the decision
ALLOW, REFUSE, or CLARIFY
Every response includes a structured decision, a reason code, and a receipt ID tied to an immutable audit chain. ALLOW means proceed. REFUSE means halt. CLARIFY means ask for more before proceeding.
03 — Enforce it
Block on REFUSE. Prompt on CLARIFY.
Your platform blocks execution on REFUSE and surfaces a narrowing prompt on CLARIFY. Log the receipt_id alongside every governed event. That's the complete audit trail — no reconstruction needed.
Guarantees

Built into the architecture.

01
Deterministic Semantics
The governance engine is not a model. It is a fixed mathematical function. Same input, same context, same policy version — identical output, always.
02
Obfuscation Resistance
L2 normalizes space-separated characters, dot-separated sequences, apostrophe variants, and Unicode substitutions before pattern matching. Obfuscation attempts are collapsed before they reach classification.
03
Fail-Closed Default
Missing field, internal error, empty input, unreachable audit sink — all produce immediate REFUSE. The system does not proceed in an unverified state.
04
Non-Bypassable Hard Locks
Certain violation categories — governance override attempts, CSAM, self-harm — carry no anchor exception pathway. No framing, context, or authorization object can override them.
05
Immutable Audit Chain
Every decision appends to a cryptographic chain. ALLOW, REFUSE, and CLARIFY all generate receipts. Chain integrity is independently verifiable — modification of any receipt invalidates all subsequent entries.
06
Session Trajectory Memory
Session state persists across turns. Slow-burn escalation — benign turn one, dangerous turn two — is detected automatically through accumulated risk trajectory, not just per-message analysis.
Verified Performance

1,000 red team scenarios.
Zero failures.

Tested across 15 attack categories including jailbreaks, roleplay bypass, obfuscated inputs, anchor abuse, forced certainty, fabrication pressure, multi-turn escalation, and benign-allow validation.

1,000/1,000 correct governance decisions. 0 false positives on clean requests.
boundary_override161/161
jailbreak81/81
roleplay_bypass80/80
forced_certainty80/80
fabrication60/60
anchor_abuse60/60
anchor_override60/60
obfuscated_jailbreak60/60
missing_anchor60/60
benign_allow98/98
context_undeclared50/50
edge50/50
ambiguous40/40
tool_escalation40/40
multi_turn20/20