ifixkart-admin/.agents/skills/planning_analysis/SKILL.md
Adithiyan bbd3a3e959 feat: initial commit — full platform snapshot
- Backend: FastAPI app with all models, services, schemas, routes
- Admin Panel: Next.js dashboard with all modules
  (CRM, POS, repairs, warranties, billing, notifications, etc.)
- Ecommerce: storefront (submodule) + product catalog
- AI Agents: .agents/ skill library for Antigravity IDE
- Docs: architecture, billing, notification gateway markdown
- Submodules: Backend/OpenWA, Ecommerce/File 2/storefront
2026-10-13 12:59:47 +05:30

2.3 KiB

name description
Planning and Analysis Triggers before any execution to analyze requirements, evaluate trade-offs (pros/cons), document plans, and obtain user approval.

Skill: Planning and Analysis

1. Purpose

Enforce a strict dual-phase workflow where the agent plans, analyzes advantages and disadvantages, produces a detailed report, and awaits explicit user approval before modifying code.

2. When to Use

Before starting any complex coding task, architectural shift, or feature implementation in the workspace.

3. Preconditions

  • Staging or source code analysis tools (Graphify) must be readable.
  • Writing access to the brain artifacts folder is verified.

4. Step-by-Step Checklist

  1. Analyze User Requirements: Deconstruct the task into functional objectives, data scopes, and potential bottlenecks.
  2. Draft Multiple Approaches: Compare at least two distinct strategies (e.g. Option A: quick implementation vs Option B: highly scalable component structure).
  3. Trade-off Analysis: Detail the advantages and disadvantages (pros/cons) of each approach regarding maintainability, latency, and code complexity.
  4. Produce Implementation Plan: Write a structured report to implementation_plan.md in the brain artifacts folder, listing proposed changes, open questions, and trade-offs.
  5. Request Feedback: Set RequestFeedback: true in the plan's metadata to block execution and await review.
  6. Refine & Re-plan: If the user provides feedback or requests changes, update the analysis and refresh the plan.
  7. Execute: Once the user explicitly approves the plan, begin step-by-step code modifications.

5. Validation Checklist

  • implementation_plan.md contains clear Pros/Cons sections for the proposed approaches.
  • No code changes are made to the codebase until user gives explicit approval.
  • Plan remains active and updated with comments based on user queries.

6. Examples

  • Planning custom billing rules: Comparing Option A (SQL views) vs Option B (SQLalchemy caching), listing pros/cons, drafting implementation_plan.md, and waiting for user input before touching database schemas.

7. Failure Cases

  • Coding immediately upon receiving a prompt without presenting trade-offs and waiting for plan approval.
  • Writing a plan that does not evaluate disadvantages or alternative solutions.