Life Memory MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| HOST | No | The host to bind for HTTP mode (e.g., 0.0.0.0). | |
| PORT | No | The port to listen on for HTTP mode (e.g., 3000). | |
| MEMORY_API_KEY | No | A secret required for any internet-facing deployment of the HTTP mode. |
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 |
|---|---|
| create_memory_profileB | Create a new private memory profile. Returns a secret owner token exactly once. Keep it private. |
| list_memory_profilesB | List profiles stored in this Life Memory installation. |
| start_memory_save_interviewB | Start the mandatory consent interview for saving a memory. This tool only succeeds if the user explicitly said exactly: "Сохрани воспоминания в архив". Starting an interview does NOT save a memory. |
| save_memoryA | FINAL save step after start_memory_save_interview. Never call directly. Requires a valid interview session created only by the exact phrase "Сохрани воспоминания в архив", required context fields, and explicit final user confirmation. |
| search_memoryB | Search memories in a local profile. Private memories require the profile owner token. |
| set_memory_sharingB | Change whether a memory profile is private, unlisted, or public and whether AI persona simulation is allowed. Requires owner token. |
| delete_memoryA | Delete one local memory entry. Requires owner token. This cannot remove copies already published or mirrored elsewhere. |
| export_memory_archiveC | Export a portable JSON memory archive. Public-only export is suitable for sharing or publishing. |
| discover_public_memoriesC | Find public memory archives listed in the Life Memory registry. |
| import_public_memoryB | Import a public/unlisted Life Memory archive from a URL after verifying its archive format and integrity hash. |
| load_memory_personaB | Load a consented public/imported memory profile as context for a clearly-labeled AI simulation. The host AI should then converse using only this archive, never claiming to literally be the person. |
| life_memory_install_infoB | Return canonical installation, save-consent, and safety information for Life Memory MCP. |
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 12 tools
Each tool targets a distinct resource and action: profile creation vs. memory saving vs. sharing vs. export/import. The descriptions explicitly clarify boundaries, such as start_memory_save_interview not saving a memory and save_memory being the final step.
Most tools follow a clear verb_noun snake_case pattern (list_memory_profiles, create_memory_profile, search_memory). One outlier is life_memory_install_info, which is a noun phrase rather than verb_noun, and start_memory_save_interview is slightly awkward but still readable.
With 12 tools, the set is well-scoped for a memory management server covering profiles, sharing, public registry, and consent workflows. Each tool appears to earn its place without excessive redundancy.
Core operations like create, read (search/list), delete, and sharing are present, but obvious gaps exist: no update_memory to edit an existing memory, no delete_memory_profile to remove a profile, and no list_memories to enumerate all entries in a profile without search.