phase-search

Search project files and memory using flow.md as an index.

6|Updated Mar 8, 2026
One-click install
npx skills add https://github.com/stanislavjiricek/neuroflow --skill phase-search
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: phase-search
Source: https://github.com/stanislavjiricek/neuroflow/tree/main/skills/phase-search
Command: npx skills add https://github.com/stanislavjiricek/neuroflow --skill phase-search

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a rapid and efficient way to search through project files and internal memory, ensuring you find relevant information quickly without unnecessary deep dives.

Core Features & Use Cases

  • Scoped Searching: Search specifically within .neuroflow/ (memory) or the broader project directory.
  • Index-Driven: Utilizes flow.md files as an index for faster, targeted searches.
  • Concise Summaries: Returns brief, actionable summaries of search findings.
  • Use Case: When you need to quickly locate documentation related to a specific experimental phase or find a configuration file for a particular agent, this skill helps you pinpoint it efficiently.

Quick Start

Search for files related to 'data processing' within the project's memory.

Frequently Asked Questions about phase-search

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

FAQPage Schema
How do I search project files quickly without scanning the entire codebase?

Scoped project files search uses flow.md as an index to efficiently locate relevant files and provide concise summaries, prioritizing index consultation over blind file scanning.

What is the best way to locate configuration files for a specific agent?

Locating configuration files for a specific agent is done by scoping searches to the project directory and consulting flow.md indexes to pinpoint relevant files efficiently.

Can I search specifically within internal memory for experimental phase documentation?

Searching internal memory for experimental phase documentation is supported by scoping searches to the memory tag, targeting the .neuroflow/ directory to retrieve relevant files.

Does file retrieval rely on an existing index to work?

File retrieval relies on flow.md files acting as an index to ensure rapid information retrieval in agentic workflows, avoiding unnecessary deep dives into project files.

What distinguishes index-driven codebase navigation from blind file scanning?

Index-driven codebase navigation consults flow.md files to target searches and provide concise summaries, whereas blind file scanning lacks scoping and retrieves information inefficiently.