Library of Context
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
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": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| library_shelveA | Persist one concise book of context to the local Library. Use for durable decisions, constraints, findings, and user-approved memories—not transient chatter or secrets. |
| library_shelve_documentA | Chunk and persist a larger local text document as multiple books. Set replace_edition to replace prior chunks from the same source. |
| library_consultA | Search the off-desk Library with hybrid vector, lexical, importance, and recency ranking without changing the reading desk. |
| library_desk_refreshA | Replace a session's bounded reading desk with books relevant to the current subject. Returns the prompt-ready context plus swapped-in, swapped-out, and retained IDs. |
| library_desk_getA | Read the latest bounded reading desk snapshot for a session. |
| library_desk_watchB | Refresh a reading desk immediately and periodically thereafter for a long-running task. A later focus change should call library_desk_refresh or start this tool again. |
| library_desk_stopA | Stop periodic refresh for a reading-desk session. The current snapshot remains available. |
| library_message_recordB | Shelve one conversation message for a virtual-context session. External model gateways should record assistant replies after each response. |
| library_prompt_buildC | Build a complete stateless model-input envelope from bounded recent turns plus a replacement reading desk. Full history remains shelved on disk instead of growing the live prompt. |
| library_context_prepareB | Durably append a user turn, then return the complete bounded semantic-paging envelope for the next model call. Send only the returned messages to the model. |
| library_context_commitB | Durably append the assistant or tool result after a governed model call. The outbox indexes it asynchronously while the recent ring makes it immediately visible. |
| library_context_protectA | Persist critical instructions, decisions, active plans, or unresolved state as protected context that remains eligible for every governed prompt until released. |
| library_context_releaseA | Release one protected event so it can be paged normally. The durable event is retained. |
| library_context_statusB | Read context-governor watermarks, queue occupancy, prompt pressure, and worker health for one agent thread. |
| library_context_flushC | Wait until this thread's durable outbox has been embedded and indexed through its recorded watermark. |
| library_statsA | Inspect Library tier health, usage, cache hit/miss counters, and embedding configuration. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/hwillGIT/library-of-context'
If you have feedback or need assistance with the MCP directory API, please join our Discord server