A Guide to Automating Your Proposal Workflow with a Command-Line AI

How to Supercharge Your Proposal Workflow with a CLI AI

The proposal process. It’s a necessary beast in business development, but the cycle of digging through old documents, finding winning case studies, and stitching it all together into something new and compelling was a total momentum killer. What if you could build a better proposal, faster, without ever leaving the cozy, focused environment of your command line?

It turns out, you can. By treating my past wins as a knowledge base and pointing a command-line AI assistant at them, I’ve completely changed the game. Here’s a look at how I’m turning my terminal into a powerful co-pilot for drafting content.

The Goal: Learn from the Past to Build the Future

The mission is straightforward: draft a killer response for a new client. But instead of staring at a blank page, I’m going to use my collection of successful proposals as the source material. The AI will do the heavy lifting, and I’ll stay in the driver’s seat.

Step 1: Just Start the Conversation

My terminal isn’t just for Git commits and server commands anymore. It’s where the creative work starts. First things first, I fire up the AI assistant.

ai-tool-name

With that one command, the session is live. The AI is now aware of my file system and ready for instructions.

Step 2: Point to the Treasure

My best proposals are the blueprints. I need to guide my AI co-pilot to the right folder. It’s as simple as using standard commands, and the tool is smart enough to follow along, even correcting my occasional typo.

User: cd deskktop

AI: I think you meant 'Desktop'. Running 'cd Desktop'...

Then, I navigate right into the heart of my past work.

cd 'Project-Docs/Proposals/Internal'

We’re now in the directory that will serve as our single source of truth for this task. This is our context.

Step 3: A Quick Look Around

Before I ask the AI to start analyzing, I like to get a sense of the landscape myself. The ls -F command gives me a quick list of the files, confirming the exact source material it has access to.

ls -F

The output gives me a clear view:

ClientA_Website_Design_Proposal.pdf

ClientB_Brand_Strategy_Proposal.pdf

ClientC_Marketing_Services_Proposal.docx

…etc

Perfect. Now the real magic begins.

Step 4: Generate the Blueprint

This is the moment that feels like a superpower. Instead of me trying to synthesize the structure of my best work, I delegate it. I ask the AI to create a high-level outline based on the common threads in all those documents.

My prompt is direct and conversational:

“Read these files and generate a table of contents for a new proposal based on their common structure.”

The AI processes the documents and, in seconds, produces a solid, proven framework.

Introduction & Executive Summary

Understanding the Challenge

Our Proposed Solution

Phase 1: Discovery & Strategy

Phase 2: Design & Development

Phase 3: Launch & Support

Project Timeline

Team & Expertise

Pricing & Investment

Next Steps

Gemini CLI used in Warp

Just like that, I have a comprehensive skeleton for the new proposal, built from the DNA of my previous successes.

Step 5: Draft the Story

With the structure locked in, it’s time to flesh it out. I can ask the AI to draft the whole thing at once, but I prefer to maintain more control by tackling it section by section. The prompt ties everything we’ve done so far together:

“Awesome. Now, using that structure, help me draft a response for the new client proposal.”

The AI gets to work, generating content that adopts the tone, style, and key details from my best past work, all perfectly organized into the new structure it just created for me.

A Smarter, More Focused Workflow

In just a few minutes, my terminal transformed from a code-execution tool into a content-generation engine. By starting with a structured outline first and then drafting the content, I ensure a logical flow while leveraging my own best work as the foundation.

This command-line approach isn’t just about speed; it’s about staying in the zone. By remaining in a focused, keyboard-driven environment, I minimize distractions and keep my creative momentum flowing. I’m not just writing proposals faster — I’m building them smarter.

Tools Used:

Gemini CLI

Claude Code

Warp