Skip to main content
POST
/
contexts
/
{context_name}
/
resolve
curl -X POST \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "scopes": {
      "user_id": 123,
      "project": "web-app"
    },
    "query": "user authentication issues"
  }' \
  https://api.contextbase.dev/v1/contexts/user-interactions/resolve

{
  "content": "USER CONTEXT: User ID 123 has previously reported login issues on 2024-01-15. Recent interactions show confusion about password reset flow. User preferences: technical explanations, quick solutions.\n\nINTERACTION HISTORY: User asked about password reset but seemed confused about email verification step. Frustration level: moderate. Previous attempts: 3.",
  "metadata": {
    "token_count": 847
  }
}

​
Overview

The context resolve endpoint fetches compressed, data-rich content from a single context. This is useful when you only need content from one specific context source, rather than combining multiple contexts together. For combining multiple contexts in a specific order, use the prompt resolve endpoint.

​
Path Parameters

​
context_name
string
required
Name of the context to resolve (e.g., “user-interactions”, “knowledge-base”)

​
Request Body

​
scopes
object
Scoping parameters to personalize the context. These filter the context to include only data relevant to specific users, projects, or environments.
​
query
string
Optional query string for RAG contexts. This helps filter and rank relevant content, preventing context rot by including only information semantically similar to your query.

​
Response

​
content
string
required
The compressed context content, optimized for prompting your LLM with relevant contextual awareness from this specific context.
​
metadata
object
Additional metadata about the resolved context.

​
Example Request

curl -X POST \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "scopes": {
      "user_id": 123,
      "project": "web-app"
    },
    "query": "user authentication issues"
  }' \
  https://api.contextbase.dev/v1/contexts/user-interactions/resolve

​
Example Response

{
  "content": "USER CONTEXT: User ID 123 has previously reported login issues on 2024-01-15. Recent interactions show confusion about password reset flow. User preferences: technical explanations, quick solutions.\n\nINTERACTION HISTORY: User asked about password reset but seemed confused about email verification step. Frustration level: moderate. Previous attempts: 3.",
  "metadata": {
    "token_count": 847
  }
}

​
Use Cases

  • Single-Source Context: Get context from one specific data source (user interactions, knowledge base, etc.)
  • Focused AI Responses: Generate responses based on one type of context without mixing data sources
  • Context Testing: Test individual contexts before combining them in prompts
  • Lightweight Applications: Simple use cases that don’t need multiple context sources

​
How Context Resolution Works

  1. Data Collection: All published data within the context is gathered
  2. Scope Filtering: Only data matching your scopes is included (e.g., specific user’s data)
  3. RAG Filtering: If query provided, only semantically relevant content is selected
  4. Content Compression: The context content is compressed and optimized
  5. Token Limiting: Result is trimmed to respect context token limits
  6. Structured Output: Final compressed context ready for your LLM prompts

​
When to Use Context vs Prompt Resolve

  • Use Context Resolve: When you need content from a single, specific context
  • Use Prompt Resolve: When you need to combine multiple contexts in a specific order
  • Single contexts: User interactions, knowledge base articles, system logs
  • Multiple contexts: Complete AI prompts combining user data + knowledge + system info