suboss87/FDEOps

What it solves

FDEOps provides a structured framework for engineers and consultants working on-site with clients to manage the non-coding aspects of AI delivery. It solves the problem of tracking client briefs, decision logs, delivery evidence, and handover processes that typically live in fragmented notes or separate repositories, ensuring that what was promised is actually measured and accepted by the client.

How it works

The system operates as a set of "skills" that can be integrated into AI coding agents (like Claude Code or Cursor). It creates a dedicated, local-only engagement memory folder (.fde/) containing plain Markdown files for each client.

When a user describes a situation to the agent (e.g., "I just finished a meeting"), the agent uses the @fde skill to identify the correct workflow—such as /brief, /plan, or /ship—and guides the user through the necessary steps to produce a record. The CLI provides a read-only HTML dashboard (the "fieldbook") to visualize risks, next actions, and delivery gaps across multiple clients.

Who it’s for

Forward deployed engineers, independent consultants, and small agencies who manage multiple client engagements and need a rigorous way to track commitments, evidence of success, and operational handover.

Highlights

• 30 Specialized Skills: Pre-defined workflows for every stage of engagement, from initial landing and discovery to shipping and final handover.
• Local-First Memory: Client records are stored as plain Markdown files on the user's laptop, ensuring data ownership and privacy.
• Agent Integration: Works as a plugin for Claude Code or via adapters for Cursor, allowing the AI to help maintain the engagement record.
• Evidence-Based Delivery: Explicitly separates promises from measurements and customer acceptance to prevent scope creep and ensure successful project closure.
• Portfolio Dashboard: A CLI-driven HTML dashboard to track the status and only the most critical items across all active clients.