Copilot Memory MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| COPILOT_MEMORY_DB | No | Custom database location. By default, memories are stored in ~/.copilot-memory/memory.db. | ~/.copilot-memory/memory.db |
| COPILOT_MEMORY_MAX_MONITOR | No | Maximum monitoring duration in seconds. Default is 1 hour (3600). | 3600 |
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
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| save_memoryA | Save a piece of knowledge to persistent memory. |
| recall_memoriesA | Search and recall past memories. |
| update_memoryA | Update an existing memory. |
| forget_memoryB | Delete a memory that's no longer relevant. |
| memory_statsA | Get statistics about stored memories. Call this to understand what's in your knowledge base. |
| monitor_commandA | Run a shell command repeatedly and collect output over time. |
| watch_fileA | Watch a file for changes or a specific pattern. |
| poll_urlA | Poll a URL repeatedly until it returns expected status/content. |
| run_long_commandA | Run a long-running shell command and capture its full output. |
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 9 tools
The memory tools (recall, save, update, forget, stats) are clearly distinct, and the monitoring tools (monitor_command, run_long_command, watch_file, poll_url) each target a different mechanism. The only potential confusion is between monitor_command and run_long_command, but their descriptions explicitly differentiate repeated polling from single long-running execution.
Most tools follow a verb_noun convention (save_memory, update_memory, monitor_command, watch_file, poll_url, run_long_command). Minor deviations include recall_memories (plural noun) and memory_stats (noun_noun), which break the strict pattern but remain predictable and readable.
At 9 tools, the count is reasonable and not overwhelming. However, the set mixes two unrelated domains (memory persistence and process/URL monitoring), making the scope feel broader than the server name suggests. Still, no tool is redundant and the number is appropriate for its combined purpose.
The memory lifecycle is complete with recall, save, update, delete, and stats. The monitoring tools cover common asynchronous scenarios: repeated command polling, single long-running command, file watching, and HTTP polling. Minor gaps exist (e.g., no direct tag-based listing, no explicit 'wait' tool), but agents can work around these.