investigate-before-recommend
CommunityStop AI drift by investigating first.
Authormealpac
Version1.0.0
Installs0
System Documentation
What problem does it solve?
Prevent AI drift by enforcing a structured investigation process before making recommendations about a user's existing systems or infrastructure. This approach ensures recommendations are grounded in reality and aligned with what actually exists.
Core Features & Use Cases
- Investigation-first workflow: establish available investigation tools (MCP servers, database access, file system, browser, GitHub) and document findings.
- Step-driven protocol: systematically identify tools, investigate, ask clarifying questions, verify assumptions, document current state, and ground recommendations in verified reality.
- Safety & quality controls: flag potential drift early with red flags and ensure grounded, non-generic recommendations.
Quick Start
Start by listing available investigation tools and then follow Step 2 to investigate existing infrastructure before making recommendations.
Dependency Matrix
Required Modules
None requiredComponents
references
💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: investigate-before-recommend Download link: https://github.com/mealpac/manus/archive/main.zip#investigate-before-recommend Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 510,000+ vetted skills library on demand.