EDGECASTS
← Back to library
intermediate Operations Reports 21:07 · Apr 9, 2026

Weekly Reports That Write Themselves

Hand a week of raw project activity to an AI and get back a one-page update that sounds like you wrote it. The Sunday-night report ritual ends this Sunday.

● Coming this week

EPISODE UPLOADING

Recording is in the can; we're cutting and uploading it now. Chapters, prompts, and the full transcript are below; read ahead and we'll drop you an email the moment the video lands.

Get the drop notification →

Steal these

THE PROMPTS

One click and they’re on your clipboard. Paste into Claude, ChatGPT, Gemini. Whatever you pay for. Run the workflow you just watched.

01

Activity → narrative report

Claude Sonnet 4.5
You are drafting a weekly executive update from raw project activity. Output a one-page markdown report with exactly these sections: Headline, Shipped, Slipped, In flight, Stuck, Numbers. Tone: plain, direct, written for a busy reader. No "moving forward", no "leveraging", no "stakeholders". Active voice.

ACTIVITY DATA:
<<paste the JSON/CSV of issues, PRs, messages here>>
02

Voice calibration

Claude Sonnet 4.5
You wrote this draft report for me, but it doesn't sound like me. Below are three genuine samples of my writing. Rewrite the draft so a colleague who knows my writing couldn't tell which was machine-generated. Preserve all facts and numbers exactly.

MY VOICE SAMPLES:
<<paste three of your own past paragraphs here>>

DRAFT TO REWRITE:
<<paste the draft here>>
03

Decisions, risks, and asks

GPT-5
Below is a machine-generated weekly report on my team's work. It describes what happened. That is only two-thirds of the job. Also propose three items in each category: DECISIONS my team made, RISKS emerging, ASKS I have. Be specific, name names. Do not invent risks.

REPORT:
<<paste the generated report here>>

Use these anywhere. If they end up in a client deck, "borrowed from EdgeCasts" in the footer is how we keep the lights on.

Full transcript +
You are not a staff writer. Your only job on the weekly report is the decisions section. Everything above it should be generated from the tools you already use. Most weekly reports spend 80% of their words recapping activity that is already logged somewhere: in Linear, in GitHub, in Slack, in the CRM. That recap is the highest-leverage thing to automate. A voice-calibration pass against three of your own past paragraphs turns the draft into something your team won't recognize as machine-written.

You are not a staff writer. Your only job on the weekly report is the decisions section. Everything above it should be generated from the tools you already use.

The architecture

Three prompts, chained, with you at the end: activity → narrative, voice calibration, and decisions & risks. You edit the last section. Ship.