PersonaMCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| PERSONAMCP_HOME | No | Selects an alternate data directory location. The default data directory comes from your operating system's application-data location. This can also be specified with --home /private/path. |
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 |
|---|---|
| get_style_profileB | Return measured outgoing-message style for a global or user-assigned context. |
| get_person_styleA | Get one-to-one writing statistics for an exact recipient name, never group history. |
| search_messagesA | Search owner's messages; use recipient/platform filters to minimize disclosure. |
| find_similar_interactionsC | Return bounded real incoming/reply examples, marking semantic/lexical coverage. |
| get_writing_contextC | Return compact style and up to three untrusted examples; never generate a reply. |
| get_persona_summaryA | Return compact profile and index coverage, without raw conversation history. |
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 6 tools
Several tools retrieve style/profile data (get_person_style, get_style_profile, get_persona_summary, get_writing_context) with subtle scope differences; descriptions clarify recipient vs global vs compact context, but an agent could still misselect. search_messages and find_similar_interactions are more distinct.
Names use snake_case and mostly verb_noun form (get_*, search_*, find_*), but the mix of get/search/find and slight noun-order variation (get_person_style vs get_style_profile) keeps it from perfect consistency.
6 tools is well within the ideal range and each seems to cover a distinct retrieval need, though some overlap exists. No excessive tool bloat.
The set covers style retrieval, persona summary, message search, similar examples, and writing context, which is solid for a read-only persona/writing assistant. Minor gaps include no explicit index/refresh or persona management operations, but likely outside intended scope.