Engineering
explore avatar

explore

Meta-skill for structured, multi-depth codebase exploration including architectural analysis, fast structural overviews, and deep-dive documentation workflows.

Introduction

Explore is a comprehensive meta-skill designed for developers and software agents operating within the Continuous-Claude environment. It provides a read-only, intent-driven workflow to parse, analyze, and document internal codebases without modifying source files. By offering three distinct depth levels—quick, deep, and architecture—it allows users to calibrate the level of investigative detail based on their current situational needs, whether that is a rapid orientation or a complex architectural dependency map.

  • Multi-depth exploration modes: Quick (tldr structure), Deep (onboard, research, and documentation), and Architecture (call graphs and layer mapping).

  • Intent-based interactive flow: Utilizes guided questioning to determine exploration goals, focus areas, and output formats (such as documentation or implementation handoffs).

  • Context-aware assistance: Integrates with the broader Continuous-Claude agent ecosystem, including scout agents for pattern analysis and tldr-explorer tools for semantic codebase navigation.

  • Automated documentation generation: Capable of producing ready-to-use reference documents or handoff files formatted for follow-on coding agents.

  • Token-efficient analysis: Leverages semantic indexing and tldr-based abstractions to avoid unnecessary context pollution, focusing only on the relevant sections of the codebase.

  • Use this skill for initial codebase onboarding, preparing for major feature development, or investigating architectural bottlenecks.

  • Input parameters include depth levels and options such as --focus, --output (doc/handoff), and --entry for targeting specific functions or entry points.

  • This is a read-only tool; it does not perform code edits but creates the necessary context for subsequent operations.

  • Ideal for users who need to map out dependencies, understand complex modules, or maintain continuity across multi-session projects.

  • Designed for the Continuous-Claude ecosystem; requires access to internal memory caches and standard file structure analysis tools.

Repository Stats

Stars
3,747
Forks
289
Open Issues
43
Language
Python
Default Branch
main
Sync Status
Idle
Last Synced
Apr 30, 2026, 04:14 PM
View on GitHub