ContextBuilder (ctx)
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| OPENAI_API_KEY | No | OpenAI API key for LLM-powered context structuring and extraction. | |
| ANTHROPIC_API_KEY | No | Anthropic API key for LLM-powered context structuring and extraction. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| ctx.refresh.app_sourcesC | Refresh and rebuild app context state + graph for a given app. Fetches public web sources, extracts structured content, builds context graph. |
| ctx.push.starter_contextC | Push a compact starter context bundle for an app. Returns product overview, key concepts, constraints, and a navigation map for deeper slices. |
| ctx.pull.context_sliceC | Pull a targeted context slice for an app based on intent, budget, and required sections. Uses graph-based retrieval and observation masking. |
| ctx.get.app_state_summaryB | Get a high-level summary of app context state including refresh status, available sections, and graph stats. |
| ctx.get.provenanceB | Get provenance details for a delivered context bundle or slice. Returns source snapshot references and statement-level provenance. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 5 tools
Each tool has a clearly distinct purpose: get.app_state_summary retrieves status metadata, get.provenance provides source tracking, pull.context_slice fetches targeted content, push.starter_context delivers initial context, and refresh.app_sources rebuilds the underlying data. There is no functional overlap or ambiguity between these operations.
All tools follow a consistent verb.object pattern with dot notation (e.g., get.app_state_summary, pull.context_slice). The naming is uniform across all five tools, using clear action verbs (get, pull, push, refresh) paired with descriptive nouns.
With 5 tools, this server is well-scoped for context management. Each tool serves a specific role in the lifecycle (retrieval, delivery, refresh), and there are no extraneous or missing tools for the apparent domain of building and managing application context.
The tool set provides complete coverage for context lifecycle operations: retrieving state and provenance (get), pulling targeted slices (pull), pushing starter bundles (push), and refreshing sources (refresh). This covers core workflows from initialization to updates without obvious gaps.