agentmemory-mcp-tools

Indexes all 54 agentmemory MCP tools with parameters and usage guidance.

3|Updated Oct 25, 2025
One-click install
npx skills add https://github.com/Prismaibrowser/prismspace-web --skill agentmemory-mcp-tools-prismaibrowser
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agentmemory-mcp-tools
Source: https://github.com/Prismaibrowser/prismspace-web/tree/main/.agents/skills/agentmemory-mcp-tools
Command: npx skills add https://github.com/Prismaibrowser/prismspace-web --skill agentmemory-mcp-tools-prismaibrowser

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When working with the agentmemory MCP server, it is hard to know which of its 54 tools to call, what each tool does, and which parameters are required. This Skill provides a complete, source-generated map of every agentmemory MCP tool so you can pick the right tool and call it correctly the first time. ## Core Features & Use Cases - Tool Selection Guidance: Organizes tools into families (capture, retrieve, sessions, knowledge graph, slots, governance) so you can quickly find the narrowest tool for a task. - Parameter Reference: Points to an auto-generated REFERENCE.md table listing every tool's parameters, required fields, and core-set membership, generated from the tools registry so it never drifts. - Use Case: You want to save an insight from the current session and find it later. The Skill tells you to call memory_save with content, concepts, and files, then retrieve it with memory_smart_search using hybrid BM25 plus vector plus graph-expanded search. ## Quick Start Ask the agent which agentmemory MCP tool to use for saving and later recalling a session insight, and have it look up the exact required parameters before calling.

Frequently Asked Questions about agentmemory-mcp-tools

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I choose which agentmemory MCP tool to call?▼

Pick the narrowest tool for the task: use memory_smart_search for open recall, memory_recall for focused queries, and memory_sessions for session listings. The Skill groups all 54 tools into families like capture, retrieve, and governance to guide selection.

How do I save and recall memories with agentmemory MCP tools?▼

Call memory_save with content, comma-separated concepts, and file paths to store an insight. Retrieve it later with memory_smart_search, which runs hybrid BM25 plus vector plus graph-expanded search over past observations.

What is the difference between the core and full agentmemory tool sets?▼

agentmemory exposes 54 MCP tools, of which 8 are in the lean core set loaded with --tools core or AGENTMEMORY_TOOLS=core. The remaining tools load with --tools all, which is the default configuration.

Where do I find required parameters for an agentmemory tool?▼

Required parameters are listed in REFERENCE.md, an auto-generated table marking required fields with an asterisk. It is generated from the tools registry via npm run skills:gen, so it stays in sync with the source.

Can I use agentmemory tools over HTTP instead of MCP?▼

Yes, agentmemory provides REST API equivalents for the MCP tools, documented in the companion agentmemory-rest-api skill. REST handlers whitelist documented fields and silently drop unknown ones.