
Review permissions and data flows
in AI applications
We assess input handling, tool permissions and supply chain risk in internal chatbots, retrieval-augmented generation (RAG), AI agents and external models. Discuss the scope and deliverables with us. The interactive demo uses illustrative scenarios.
In operationWe assess prompt injection, MCP tool descriptions, excessive tool permissions, and supply-chain packages through representative scenarios. The interactive demo is a simulation. Service standards
Service scope and process
Dedicated analysis nodes for each customer
Each customer’s analysis runs on dedicated nodes. Customers use separate compute resources and storage.
Every decision has a record of the supporting evidence
A single record identifies the engine behind the decision, the logs it relied on, and the action it recommended. You can submit that record as audit evidence.
Past decisions can be rerun
We record the model and policy versions used for each decision. If you question a decision from last month, you can rerun it on the same data and compare the results.
People make the final decision
AI classifies and summarizes alerts. Analysts approve response actions, including server isolation and account lockouts, before they run.
AI security assessment services
LLM application assessment
We assess chatbots, RAG systems, and internal AI services against 20 criteria.
Learn moreAI agent and MCP security assessment
We review each agent’s tool permissions, delegation relationships, and MCP connections.
Learn moreModel and supply chain assessment
We assess the security of models, datasets, and plugins from external sources.
Learn moreLLM application assessment
The assessment covers 20 criteria based on the OWASP LLM Top 10, agent-specific threats, and Korean regulatory requirements. We assess safeguards against risks relevant to your service architecture. The playground is a simulation.
Key capabilities
20 assessment criteria
The assessment builds on the OWASP LLM Top 10 and adds agent-specific threats and Korean regulatory criteria.
Isolated test environment
Test inputs and results stay within the authorized environment.
Attach evidence from the reproduced finding
For each finding, the report includes the test input and the system’s observed response.
Proposed detection and prevention rules
We recommend guardrails and filter rules for each vulnerability we find.
Key assessment criteria (20)
| Classification | Assessment criteria | Description | Risk level |
|---|---|---|---|
| Input | Direct prompt injection | System instructions overridden or roles altered | High |
| Input | Indirect prompt injection | External content treated as instructions | High |
| Input | Jailbreak | Responses that circumvent safety policies | Medium |
| Input | Multilingual and encoding-related filter bypass risks | Filter bypass through Korean wording variants, Base64, and similar inputs | Medium |
| Data | System prompt disclosure | Exposure of internal instructions or business logic | Medium |
| Data | Sensitive information exposure | Personal information or trade secrets disclosed in responses | High |
| Data | RAG access control bypass risk | Search results expose documents outside the user’s access rights | High |
| Data | Training data extraction risk | Verbatim disclosure of memorized training data | Medium |
| Agent | Excessive tool permissions | Connections to tools and APIs with excessive permissions | High |
| Agent | Tool call tampering | Unauthorized email or database access triggered by prompt injection | High |
| Agent | MCP server impersonation | Connection to a malicious or tampered MCP server | High |
| Agent | Approval control bypass risk | Required human approval bypassed | Medium |
| Output | Improper output handling | Model output executed directly as code or a query | High |
| Output | Harmful content generation | Content that violates policy | Medium |
| Output | Misinformation from hallucinations | Fabricated claims or links that do not exist | Low |
| Supply chain | Model file tampering | Malicious serialized files and backdoored models | High |
| Supply chain | Data poisoning | Fine-tuning and RAG data poisoning | Medium |
| Operations | Resource exhaustion | Excessive token use or calls can drive up costs or disrupt service | Medium |
| Operations | Missing logs and audit trails | Prompts and tool calls go unlogged | Low |
| Governance | Regulatory alignment | Unmet requirements under Korea’s AI Basic Act or Personal Information Protection Act | Low |
Process
Threat modeling
We map the service architecture and data flows
- Threat model
Automated attack simulation
We run automated checks of prompt safeguards
- Automated assessment results
Reproduction
We reproduce a scenario for each criterion and record the results
- Vulnerability list
Defense design
We propose guardrails and changes to tool permissions
- Defense guide
Revalidation
We verify the fixes with follow-up testing
- Remediation verification report
Frequently asked questions
Do I need to send the model outside my environment?
We assess systems in your environment or in an isolated Research Lab. Test data stays within that environment.
Can this support work under Korea’s AI Basic Act?
The report flags findings related to risk management requirements for high-impact AI.
Explore related articles and resources.
AI agent and MCP security assessment
We inventory the tools, APIs, and MCP servers each agent can call, and record their permissions.
We check whether instructions embedded in external content can cause unauthorized actions, such as sending email or accessing a database.
Key capabilities
Permission map
We provide a one-page overview of which tools and data each agent can access.
Delegation analysis
We trace how permissions pass when one agent delegates to another.
MCP server assessment
We verify the source, version, and signature of connected MCP servers.
Tool call monitoring rules
We document normal tool call patterns and provide detection rules for deviations.
Process
Inventory
We inventory agents, tools, and MCP connections
- Permission map
Attack
We test whether tool calls can be manipulated
- Vulnerability list
Improvements
We identify which permissions to narrow and recommend changes for each one
- Recommended changes to permissions
Monitoring
We apply behavioral detection rules
- Detection rules
Explore related articles and resources.
Model and supply chain assessment
We inspect model files, datasets, and plugins from public repositories for malicious code and contaminated data.
Key capabilities
Model file inspection
We examine serialized model files, such as Pickle files, for code that could run when loaded.
Build an AI-BOM
We document the models, datasets, and libraries in use in a maintainable inventory.
Execution in an isolated environment
We run the model in the isolated Research Lab and check for unusual behavior.
Establish a model intake process
We help you establish an internal process to verify provenance and integrity before approval.
Process
Inventory
We inventory the models and data in use
- AI-BOM
Static analysis
We look for hidden code in files
- Inspection results
Runtime assessment
We test in an isolated environment
- Evaluation results
Policy development
We design a process that limits use to approved models
- Model intake policy
Explore related articles and resources.

Discuss your security needs
Tell us about your operating environment and the security scope you need, and we’ll walk you through applicable configurations.



