acp-investigate

Investigate code read-only with subagents and evidence-based reporting.

29|12|Updated May 4, 2026
One-click install
npx skills add https://github.com/bravekingzhang/agent-coding-playbook --skill acp-investigate
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: acp-investigate
Source: https://github.com/bravekingzhang/agent-coding-playbook/tree/main/skills/acp-investigate
Command: npx skills add https://github.com/bravekingzhang/agent-coding-playbook --skill acp-investigate

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill assists in understanding, explaining, tracing, auditing, or exploring code without the risk of unintended changes or fixes.

Core Features & Use Cases

  • Read-Only Exploration: Safely investigate code without altering it, ideal for learning and audit purposes.
  • Task Delegation: Utilize the Explore subagent for efficient code search and analysis.
  • Accurate Explanation: Build comprehensive, evidence-based explanations using file paths, observations, and inferences.
  • Structured Reporting: Delivers organized answers and maintains an audit trail of what was investigated.

Quick Start

Use the acp-investigate skill to trace the request flow in the 'orders' API.

Frequently Asked Questions about acp-investigate

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

FAQPage Schema
How do I perform a read-only code investigation without modifying the source?

Read-only code investigation analyzes existing logic without altering files by utilizing subagents for efficient searching. It maintains an evidence-based approach using file paths and observations to build comprehensive explanations safely.

What is the best way to trace API request flows for an audit trail?

Tracing API request flows for an audit trail is best handled by delegating search tasks to an Explore subagent. This approach maintains a structured reporting format that records exactly what was investigated.

Do I need Python to run a code exploration and performance analysis?

Yes, Python is required to run code exploration and performance analysis. The investigation process requires Python and other tools to access file content and execute commands needed for searching.

Does read-only code audit work for understanding existing code logic in large projects?

Yes, read-only code audit works for understanding existing code logic by utilizing subagents for efficient searching across large projects. It builds accurate explanations using structured evidence-based reporting.

Can I use code exploration to generate performance analysis without risking unintended fixes?

Yes, you can use code exploration to generate performance analysis without risking unintended fixes. The investigation operates strictly in a read-only mode to ensure no unintended changes occur.