How does Memory Query differ from standard Claude context?

While standard context is limited to the current active session, Memory Query allows Claude to programmatically search through a persistent database of all past interactions within a specific project.

Can I search across different projects?

No, memory stores are maintained on a per-project basis to ensure data isolation and prevent cross-context contamination.

How do I check if my memory search is working?

You can use the 'memory-daemon status' command to check the service health and 'retrieval status' to see which search layers are currently active.

A Grip ID is a unique identifier (e.g., grip:timestamp:ulid) used to precisely locate and expand specific conversation excerpts for deeper context.

What are the retrieval tiers?

Tiers are search strategies ranging from Full (Topics + Hybrid) to Agentic (TOC navigation). The system automatically selects the best tier based on query intent and available indices.

Memory Query

Name: Memory Query
Author: SpillwaveSolutions

bySpillwaveSolutions

•

생산성 및 워크플로

Retrieves historical conversation context and past session data using intelligent tiered search and intent-based routing.

The Memory Query Skill empowers Claude with long-term recall by interfacing with the agent-memory system to fetch precise information from previous development sessions. It utilizes a sophisticated multi-tier retrieval engine that automatically navigates between semantic vector search, BM25 keyword matching, and hierarchical time-based browsing based on the user's query intent. This ensures Claude can reconstruct the reasoning behind past architectural decisions, locate specific code discussions from weeks prior, and maintain a continuous knowledge base across fragmented work sessions with full explainability for every result retrieved.

주요 기능

01Detailed search explainability providing transparency into confidence scores and methods used

02Granular context expansion using unique Grip IDs for precise excerpt retrieval

033 GitHub stars

04Hierarchical TOC navigation for structured time-based browsing (Year/Month/Week/Day)

05Tier-aware retrieval with automatic fallback across semantic, keyword, and agentic layers

06Intelligent query intent classification for optimized search routing

사용 사례

01Recalling specific technical requirements or implementation details discussed in previous months

02Searching for past debugging sessions to understand how a recurring issue was previously resolved

03Browsing a project's interaction history to summarize the evolution of a specific feature or module

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add spillwavesolutions/agent-memory memory-query

For use in Claude.ai and ChatGPT

주요 기능

01Detailed search explainability providing transparency into confidence scores and methods used

02Granular context expansion using unique Grip IDs for precise excerpt retrieval

033 GitHub stars

04Hierarchical TOC navigation for structured time-based browsing (Year/Month/Week/Day)

05Tier-aware retrieval with automatic fallback across semantic, keyword, and agentic layers

06Intelligent query intent classification for optimized search routing

사용 사례

01Recalling specific technical requirements or implementation details discussed in previous months

02Searching for past debugging sessions to understand how a recurring issue was previously resolved

03Browsing a project's interaction history to summarize the evolution of a specific feature or module