Why are tools like grep and find forbidden when using this skill?

Legacy search tools are inefficient for large codebases; this skill enforces indexed memory to provide agents with a more structured and comprehensive view of the code.

Which agent roles are supported by this skill?

It supports a wide range of roles including Frontend Developers, API Architects, Test Architects, Senior Code Reviewers, and Debuggers, mapping each to a specific detective skill.

How do I add this skill to my agent?

You simply reference the appropriate detective skill (e.g., code-analysis:developer-detective) in your agent's frontmatter and include the code-analysis plugin as a dependency.

Can I use this with my own custom plugins?

Yes, the skill provides a standardized mapping and integration pattern that any custom plugin agent can use to access advanced investigation capabilities.

What is the main benefit of using Cross-Plugin Detective?

It ensures all AI agents use high-performance indexed memory (claudemem) instead of slow, legacy search tools like grep, leading to faster and more accurate code discovery.

Cross-Plugin Code Detective

Name: Cross-Plugin Code Detective
Author: involvex

byinvolvex

0•

Developer Tools

Coordinates specialized detective skills across various agent roles to ensure efficient, indexed codebase navigation and analysis.

Cross-Plugin Detective serves as the central orchestration layer for codebase investigation within Claude Code. It maps specific agent roles—such as frontend developers, backend architects, and test engineers—to tailored 'detective' skills that leverage high-performance indexed memory via claudemem. By enforcing a strict 'indexed memory only' policy, it eliminates inefficient legacy search methods like grep and find, ensuring agents find implementations, trace data flow, and analyze architecture patterns with superior accuracy and speed across complex, multi-plugin environments.

Key Features

010 GitHub stars

02Context-aware skill selection for developers, testers, and reviewers

03Strict enforcement of high-performance indexed memory search via claudemem

04Unified investigation patterns across diverse codebase architectures

05Intelligent role-to-skill mapping for frontend, backend, and architect agents

06Prevention of inefficient legacy tool usage like grep, find, and rg

Use Cases

01Optimizing agent performance by transitioning from text-searching to indexed memory

02Streamlining complex architectural analysis for senior reviewer and architect roles

03Standardizing codebase discovery across a monorepo with specialized AI agents

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add involvex/involvex-claude-marketplace cross-plugin-detective

For use in Claude.ai and ChatGPT

Download Skill