people-memory-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| PEOPLE_MEMORY_ENV_FILE | No | Path to a .env file that is read automatically | |
| PEOPLE_MEMORY_API_TOKEN | No | Optional API token to require Authorization: Bearer for the REST API | |
| PEOPLE_MEMORY_DATABASE_URL | Yes | PostgreSQL database URL for the People Memory database | |
| PEOPLE_MEMORY_DEFAULT_SOURCE | No | Default source for recorded facts | agent |
| PEOPLE_MEMORY_ENABLE_RAW_SQL | No | Whether to enable raw SQL query/write tools | false |
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 |
|---|---|
| graph_statusB | Check the graph connection and return record counts. |
| search_peopleC | Search people by name, organization, role, city, identifier, summary, or fact. |
| get_personA | Get one person's full record. Returns candidates instead of guessing when ambiguous. |
| remember_personA | Create or update a person from conversation. By default this refuses similar-name duplicates and conflicting overwrites. Ask the user, then retry with confirmed_new or overwrite only after they decide. |
| add_factC | Attach a durable fact to a person. Ask the user if the name is ambiguous. |
| record_interactionB | Record a dated call, meeting, message, meal, or other interaction. |
| connect_peopleB | Create a relationship between two existing people. Never guesses ambiguous names. |
| find_intro_pathC | Find short relationship paths from a person to anyone at a target organization. |
| stale_contactsC | List strong relationships with no recent interaction. |
| read_queryB | Run one guarded SELECT for advanced graph analysis. No DDL or data changes. |
| write_queryA | Run one guarded INSERT, UPDATE, or DELETE. UPDATE/DELETE require WHERE; DDL is blocked. |
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 11 tools
Most tools are clearly distinct (graph_status, search_people, get_person, add_fact, record_interaction, find_intro_path, stale_contacts). There is minor overlap between read_query/write_query and the higher-level person/interaction tools, since an agent could use write_query to bypass the intended workflows, but descriptions clarify their guarded, analysis-focused roles.
Tool names follow a consistent verb_noun pattern throughout: graph_status, search_people, get_person, remember_person, add_fact, record_interaction, connect_people, find_intro_path, stale_contacts, read_query, write_query. The naming is clean and predictable.
11 tools is reasonable for a people-memory graph server. Each tool covers a distinct purpose (querying, creating, interactions, relations, analysis). It's slightly on the higher end but not excessive, and no tool feels redundant.
The surface covers search, get, create/update (remember_person), add facts, record interactions, connect people, path finding, and staleness analysis. Minor gaps include no delete/forget person operation and no direct relationship update/removal tool, but core lifecycle and analysis workflows are well covered.