proactive-explorer

Gather project and environment context using Grep, Glob, Read, Bash, and WebSearch before asking clarifying questions.

425|37|Updated Mar 6, 2026
One-click install
npx skills add https://github.com/cnfjlhj/ai-collab-playbook --skill proactive-explorer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: proactive-explorer
Source: https://github.com/cnfjlhj/ai-collab-playbook/tree/main/skills/full/proactive-explorer
Command: npx skills add https://github.com/cnfjlhj/ai-collab-playbook --skill proactive-explorer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Many AI-assisted workflows stall because the assistant asks for information it could have found itself; this Skill ensures the assistant proactively gathers project and environment context so users are only asked essential, unavoidable questions.

Core Features & Use Cases

  • Automated context collection: Run fast scans (Grep, Glob) and deeper reads (Read, Bash) to locate definitions, configuration, logs, and dependencies before prompting the user.
  • Tiered exploration strategy: Use a prioritized stack—fast local searches, project CLAUDE/AGENTS knowledge, web searches, then specialized subagents—so clarifying questions are a last resort.
  • Use Case: When asked to "implement user authentication", the Skill locates existing auth code, checks package.json for dependencies, reads CLAUDE.md for past decisions, and only asks the user for unresolved policy choices.

Quick Start

Perform a proactive exploration for the task "implement user authentication" and return findings from Grep, Read, Glob, Bash, and WebSearch before asking any clarifying questions.

Frequently Asked Questions about proactive-explorer

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

FAQPage Schema
How do I stop my AI agent from asking unnecessary clarification questions during codebase exploration?

Automated context gathering reduces clarification questions by running fast scans via Grep and Glob, deeper reads via Read and Bash, and consulting project files before escalating unresolved queries to the user.

What is the best way to gather project context before implementing a new feature?

Proactively gather project context by using a tiered exploration stack: run fast local searches, read project knowledge files, perform web searches, and deploy subagents so you only ask users about unresolved policy choices.

How does automated context gathering work for bug triage in an unfamiliar codebase?

Automated context gathering for bug triage applies a tiered exploration strategy using Grep, Glob, Read, Bash, and WebSearch to locate definitions, logs, and dependencies before asking the user for clarification.

Can I use Grep and Glob to locate existing authentication code before planning a feature?

Yes, Grep and Glob perform fast local scans to locate definitions and configurations, which combined with Read and Bash, ensures the assistant checks package files and project knowledge before asking essential questions.

When should I ask the user for clarification during an agent workflow?

Ask the user for clarification only after exhausting a prioritized exploration stack of local searches, project knowledge consultation, web searches, and specialized subagents, ensuring questions are a last resort.