search

Search prior prompts and observations across sessions using MCP.

Updated Jun 2, 2026

Search Tags:#search #retrieval #memory #coding #prompts #sessions #observations

Ready to run

npx skills add https://github.com/chicago-joe/manymems --skill search-chicago-joe

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: search
Source: https://github.com/chicago-joe/manymems/tree/main/plugin/skills/search
Command: npx skills add https://github.com/chicago-joe/manymems --skill search-chicago-joe

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps users surface relevant prior work and prompts across sessions to inform current tasks and decisions.

Core Features & Use Cases

Cross-session search for prior prompts, implementations, and decisions to accelerate development and debugging.
Retrieval and synthesis of related observations to provide context and traceability.
Use Case: When planning a feature, quickly surface how similar problems were solved in earlier sessions to guide implementation.

Quick Start

Run a search over past sessions using the MCP search tool and review the top results.

Frequently Asked Questions about search

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

FAQPage Schema

How do I search across previous sessions for prior coding prompts and decisions?▼

Cross-session search uses deterministic MCP calls and fallback observation retrieval to surface prior coding prompts, implementations, and decisions for current task evaluation. You retrieve top observations and synthesize them to accelerate development and debugging.

Can I retrieve context from past sessions for multi-user scenarios?▼

Multi-user session retrieval supports surfacing prior work and context across different users. It applies deterministic search calls to find relevant observations, enabling you to trace decisions and implementations from sessions beyond your current one.

What is the best way to find how similar problems were solved in earlier coding sessions?▼

Finding prior implementations involves retrieving top observations across sessions to synthesize related context. This search capability helps you review how similar problems were solved previously to guide current feature planning and implementation.

How does memory search handle fallback when initial session retrieval fails?▼

Memory search relies on fallback observation searches when deterministic MCP search calls do not return sufficient results. It retrieves top observations across sessions to ensure relevant prior work and prompts are surfaced for synthesis.

Does cross-session memory search work without external dependencies?▼

Cross-session memory search operates without external dependencies, using built-in MCP search tools and fallback observation retrieval. You can run a search over past sessions directly and review the top results without installing additional components.

Why should I use session search instead of manually browsing past coding prompts?▼

Session search automates the retrieval of prior prompts and observations across sessions, providing traceability and context synthesis. Manually browsing past work lacks the deterministic search calls needed to efficiently surface relevant decisions for current development tasks.