Agent( )





Pattern
Tool Context: experimental_context
Inject request-scoped context (user identity, database handles, session) into ToolLoopAgent tools via experimental_context, so the model never supplies sensitive identity in tool inputs.
IntermediateJun 2026View the full pattern on desktop


Problems solved
- Keep user identity out of model-controlled tool inputs
- Share database or service handles across tools via experimental_context
- Scope tool execution to the authenticated request
- Reduce prompt-injection risk around cross-user data access
Use cases
- Per-user account assistants
- Multi-tenant agent tools
- Injecting DB or session into isolated tool calls
- Identity-scoped agent backends
AI SDK APIs
ToolLoopAgentexperimental_contexttoolstepCountIsgateway
Tags
aiagentsexperimental-contexttool-contextsecurityauthtool-loop-agentai-sdkagent-api
Requirements
- Node.js
- 24+
- Memory
- 512MB
- Other
- OpenAI API key
Environment variables
OPENAI_API_KEYGet key
External services
openai
Ready to use?
View the full interactive pattern with live preview and copy the source code.
View the pattern on desktopRelated patterns


Loop Control: stopWhen
Demonstrate ToolLoopAgent loop control with stopWhen. Covers stepCountIs, hasToolCall, combined conditions, and a custom StopCondition with step timeline visualization.
aiagents+7
View pattern β


PrepareStep: Trim Message History
Demonstrate ToolLoopAgent prepareStep for trimming long message history. Pre-seeded 24-message conversation shows context compression on the first run with per-step trim telemetry.
aiagents+6
View pattern β



