SIM swaps and number porting wait for approval.
Operators serve millions of subscribers with agents for bill disputes, plan changes, SIM renewal and fault tickets. Law 5809 and BTK regulations protect subscriber and traffic data. A SIM swap is the most critical point for fraud. Thulpar binds SIM and number-porting tools to a hold-and-approve rule in Cedar, masks subscriber PII and catches toxic language in both directions.
tr-banking-bddk
On-premise, Performance tier, replicated. Per-tenant detector pools.
- Electronic Communications Law 5809 · BTK regulations
- KVKK 6698 Art. 9, 12
- Regulation on the Processing of Personal Data and Protection of Privacy
Scale and latency
On high-volume channels the guardrail has to scale horizontally and pool per tenant.
Watch the runs →The scenarios below are derived from the product’s real rules.
Every scenario is an agent run written to the product’s real rules. It shows the stage, the rails, the decision, the finding, the approval and the ledger entry step by step.
3 scenarios →What comes up in practice
Scale and latency
On high-volume channels the guardrail has to scale horizontally and pool per tenant.
SIM-swap risk
A prompt that skips identity verification must never lead an agent to swap a SIM.
Traffic data
Location and call records have to be bounded before they reach the model.
3 scenarios
The player opens on the flagship scenario. Pick the others below.
A subscriber asks for a SIM with a prompt that says “skip identity verification”; it is denied, and even the verified request waits for four eyes.
10 GR-*
| id | control | OWASP | rails |
|---|---|---|---|
| GR-INJ-01 | Direct injection and jailbreak at the gate (TR+EN) | LLM01 / ASI01 / MCP06 | injection.direct, jailbreak |
| GR-TOOL-02 | Durable human approval on high-impact actions, four eyes | LLM06 / ASI02, ASI09 | labels.sink |
| GR-TOOL-01 | Tool-call privilege policy (default deny) | LLM06 / ASI02, ASI03 / MCP02 | tool.authz, tool.args |
| GR-PII-01 | PII and special-category detection + masking (TCKN, IBAN, phone, health) on input/tool/output/log | LLM02 / — / MCP10 | pii.detect, pii.redact, file.metadata_pii |
| GR-MEM-01 | Memory-poisoning policy + provenance | — / ASI06 / MCP10 | memory.write |
| GR-CONT-01 | Content-harm policy (AILuminate 12 + TR) with tenant thresholds | LLM09 / ASI09 | content.input, content.output, toxicity, topic, groundedness, language, image.content_safety, banlist, competitors |
| GR-LOG-01 | Tamper-evident event log (model/tool/policy/approval/hit), ≥ 6 months, auditor export | — / — / MCP08 | |
| GR-IFC-01 | Information-flow labels (integrity, confidentiality, tenant, PII class) enforced at sensitive sinks | LLM02 / ASI01, ASI03 / MCP10 | labels.sink, pii.detect |
| GR-EXF-01 | Tenant egress allowlist; no remote images in chat; no PII in URLs; SSRF | LLM02, LLM05 / ASI02 / MCP06 | exfil, tool.args, file.qr |
| GR-RES-01 | Per-run rate/cost/loop limits, circuit breaker | LLM10 / ASI08 | budget |
See it in your own environment.
A one-hour technical demo: your scenario, your policy pack, on a closed-circuit install. Then the install kit and a trial licence for a pilot.