ClawSouls MCP Server
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": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| soul_searchA | Search AI agent personas on ClawSouls by keyword, category, or tag |
| soul_getB | Get detailed information about a specific persona |
| soul_installA | Download a persona from ClawSouls and generate a CLAUDE.md file |
| soul_scanA | Run SoulScan safety verification on Soul Spec files. Analyzes persona files against 53 safety patterns and returns a grade (A+ to F) with actionable recommendations. |
| soul_rollback_checkA | Detect persona drift by comparing current Soul Spec files against their original committed versions. Returns drift severity and recommended actions. |
| memory_searchA | Search agent memory using TF-IDF ranking. Returns a compact index of matching sections with relevance scores. Use memory_detail to fetch full content of interesting results. Searches MEMORY.md + memory/*.md files. |
| memory_detailA | Fetch full content of a specific memory section. Use after memory_search to get details for high-scoring results. |
| memory_statusA | Show current status of agent memory files — list files, sizes, last modified dates, and git status. |
| memory_syncA | Sync agent memory files with a remote Git repository for multi-agent Swarm Memory. Supports init (setup), push (upload changes), pull (download changes), and status. |
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
Each tool has a clearly distinct purpose with no ambiguity. Memory tools (memory_detail, memory_search, memory_status, memory_sync) handle different aspects of memory management, while soul tools (soul_get, soul_install, soul_rollback_check, soul_scan, soul_search) each perform unique persona-related operations. The descriptions clearly differentiate their functions.
Tool names follow a perfectly consistent snake_case pattern with clear prefix organization. All memory tools start with 'memory_' and all persona tools start with 'soul_', creating predictable groupings. The verb+noun structure is maintained throughout (e.g., memory_search, soul_install).
With 9 tools, this server is well-scoped for its dual-domain purpose (memory management and persona management). The count allows comprehensive coverage without bloat, and each tool clearly earns its place in the workflow. This is an ideal number for the server's apparent scope.
The tool surface provides excellent coverage for both memory and persona domains. Memory tools support search, detail retrieval, status checking, and synchronization. Persona tools support discovery, installation, safety verification, drift detection, and information retrieval. The only minor gap is the lack of persona modification/update tools, but agents can work around this by reinstalling or using memory tools for customizations.