Before Vijil can evaluate or protect an agent, you need to register it. Registration tells Vijil how to communicate with your agent and how much visibility you’re granting for analysis.
The Agent Registry
Navigate to Agents in the sidebar to open the Agent Registry.
The registry displays all agents in your workspace with key information:
| Column | What It Shows |
|---|
| Agent Name | Identifier you assigned during registration |
| Owner | Team member who registered the agent |
| Status | Draft, Active, Under Review, or Archived |
| Access Level | Black Box, Grey Box, or White Box |
| Last Evaluated | When Diamond last ran an evaluation |
| Trust Score | Current score from most recent evaluation |
Access Levels
Vijil supports three access levels, each enabling progressively deeper evaluation. The access level you choose determines what Vijil can analyze and what vulnerabilities it can detect.
| Access Level | What You Provide | What Vijil Can Test |
|---|
| Black Box | API endpoint and credentials | Input/output behavior only |
| Grey Box | Model config, MCP config, A2A config | Vulnerabilities traced to agent composition |
| White Box | Full configuration and source code | SAST and DAST analysis for thorough audit |
Black Box
Black Box registration requires minimal information—your agent’s API endpoint and authentication. Vijil treats the agent as an opaque system, testing only its observable behavior.
This approach works well for:
- Third-party agents you don’t control
- Initial evaluations before deeper analysis
- Agents where you want to test the API contract
Required fields:
- Agent Name — A descriptive identifier
- Status — Start with Draft for initial setup
- Agent URL — Where your agent accepts requests
- API Key — Authentication credentials (you can add multiple keys)
- Rate Limit Per Minute — Matches your provider’s limits to avoid throttling
Grey Box
Grey Box expands visibility into your agent’s composition. This enables Vijil to trace vulnerabilities to specific components—whether a prompt injection succeeded because of the model choice, MCP tool exposure, or delegated agent behavior.
Additional fields:
| Section | Fields | Purpose |
|---|
| Role & Instructions | Role, System Prompt | Helps Vijil understand expected behavior |
| Model Configuration | Hub, Model Name, Parameters | Identifies model-specific vulnerabilities |
| External Resources | MCP Proxy/Server, Access Level | Tests tool-use attack surfaces |
| Delegated Agents | A2A Card URLs | Evaluates multi-agent orchestration risks |
| Hub | Models |
|---|
| OpenAI | GPT-4, GPT-4o, o1, o3 |
| Anthropic | Claude 3, Claude 4 |
| GCP | Gemini via Vertex AI |
| Bedrock | Claude, Titan, Llama via AWS |
| Groq | Llama, Mixtral via Groq |
| Together | Llama 4 via Together AI |
| Mistral | Mistral, Magistral |
| Custom | Any OpenAI-compatible API |
If your agent uses MCP tools or delegates to sub-agents, Grey Box registration enables Vijil to test the full execution graph—not just the top-level API.
White Box
White Box provides full source code access for static analysis (SAST) combined with dynamic testing (DAST). This is the most thorough evaluation path.
Additional fields:
- Source Code — Paste code directly, or leave empty to pull from repository
- GitHub Repository — URL, username, and access token for private repos
White Box evaluation can identify vulnerabilities that Black Box testing would miss—hardcoded secrets, unsafe deserialization, SQL injection in tool implementations, and similar code-level issues.
Registering an Agent
- Click + Register Agent to open the registration modal
- Choose Manual for direct entry or Import to load from a file
- Fill in the Black Box fields (required for all agents)
- Expand Grey Box if you want compositional analysis
- Expand White Box if you want source code analysis
- Click Register Agent
Your agent appears in the registry with status Draft. Change to Active when ready for evaluation.
Managing Agents
Click the More Actions menu (three dots) on any agent row to access management options.
| Action | What It Does |
|---|
| Edit | Modify agent configuration |
| Clone | Create a copy with a new name |
| Export JSON | Download configuration for backup or migration |
| Archive | Remove from active list while preserving history |
| Delete | Permanently remove agent and all associated data |
Editing an Agent
Select Edit to open the agent configuration modal with current values populated.
Make your changes and click Save Changes.
Changing an agent’s configuration may affect ongoing evaluations. If you need to test different configurations, consider using Clone to create a variant instead.
Agent Status
Agents progress through lifecycle states:
| Status | Meaning |
|---|
| Draft | Configuration in progress, not ready for evaluation |
| Active | Ready for evaluation and protection |
| Under Review | Currently being evaluated |
| Archived | Preserved for reference but inactive |
Integration Guides
For provider-specific configuration details, see the Cloud Providers guide in the Developer Guide.
Next Steps
