Guide · Agent skill
How to use the Executive Summary Writer agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The exec-summary skill is a set of instructions that lets an AI agent write a 1-2 page executive summary for a feature that is ready for development. The one document it produces is meant to orient sales, leadership and engineering at the same time. This guide explains what the skill does, how to install it, how to ask for it in plain language, and what to check before you trust the output.
Executive Summary Writermvschwarz/openrig · Apache-2.0 Download the skill Quick start
- 1Download the ZIP. Get the exec-summary ZIP from the page where you found this guide. It contains SKILL.md.
- 2Unzip the skill. Extract it into ~/.claude/skills/ for personal use, or into a project's .claude/skills/ folder for one project.
- 3Or upload it in the Claude apps. If you use the Claude apps, go to Settings → Capabilities → Skills and upload the skill there.
- 4Prepare the feature folder. Make sure the folder has SPEC.md, background.md, validation.md and supporting/mockup-ascii.md.
- 5Ask for the summary. Request an executive summary for that feature folder in plain language. The agent saves executive-summary.md in the same folder.
- 6Review before sharing. Check names, figures and claims against your source files, then share the document with sales, leadership and engineering.
What the exec-summary skill makes the agent do
When the skill is installed and you ask for a summary, the agent reads the feature folder's SPEC.md, background.md, validation.md and supporting/mockup-ascii.md. It then writes a single executive-summary.md file in that feature folder, about 600-800 words excluding mockups and FAQ.
The document is written in present tense, as if the feature already exists. It covers the problem, the solution, who it is for, why now, how it works, key decisions, visual preview, what is not being built, dependencies, success metrics and short FAQs for sales, leadership and engineering.
What you need before you start
The skill needs a feature folder that contains the files it reads. The source names four: SPEC.md, background.md, validation.md and supporting/mockup-ascii.md. The agent also needs a runtime that can read and write files in your project.
If your team does not use this folder layout, or the feature has no spec or validation files yet, the skill has little grounded material to work from. The skill's own guidance warns that the agent may then invent customers or metrics.
- SPEC.md with acceptance criteria, business rules, scope and user stories
- background.md with customer drivers, competitive context and regulatory considerations
- validation.md with demand evidence, status quo, desperate user and narrowest wedge
- supporting/mockup-ascii.md with the mockups to pull 2-3 key screens from
- An agent runtime that can read and write files in the project
feature-folder/
SPEC.md
background.md
validation.md
supporting/
mockup-ascii.mdHow to install the exec-summary skill
Download the ZIP from the page where you found this guide. The package contains one file, SKILL.md. Unzip it so the skill folder sits in one of two places: your personal skills folder at ~/.claude/skills/, or a project's .claude/skills/ folder if you want the skill to apply only to that project.
If you use the Claude apps instead of the file system, you can upload the skill under Settings → Capabilities → Skills.
- Personal install: unzip into ~/.claude/skills/
- Project install: unzip into the project's .claude/skills/ folder
- Claude apps: Settings → Capabilities → Skills, then upload
How to trigger the skill with a natural request
The skill is meant to be used when a feature is ready for development. Ask the agent in plain language to write the executive summary for a specific feature folder, and name that folder so the agent knows which files to read. The source does not list fixed trigger phrases, so describe what you want in your own words.
Example wording you could try: 'Write the executive summary for the billing-alerts feature folder.' Replace the folder name with your own.
What the output file contains
The agent saves the result as executive-summary.md inside the feature folder. The file starts with a short metadata block containing a title, the feature name, a status of ready, in-progress or shipped, a Jira ticket key and an updated date.
The body follows a fixed structure, so every summary reads the same way. Sales, leadership and engineering each get a short FAQ at the end.
- The Problem and The Solution, written in customer-readable language
- Who It's For and Why Now
- How It Works, with 5-8 user-facing capabilities
- Key Decisions, aimed at non-obvious design choices
- Visual Preview, with 2-3 curated ASCII mockups
- What We're NOT Building, Dependencies & Sequencing and Success Metrics
- FAQ for Sales, Leadership and Engineering
Safety notes: scripts and network access
This skill does not run scripts and does not use the network, according to its listing. Its work is reading and writing files inside the feature folder.
The output is only as reliable as the inputs. The skill asks for named customers, dollar amounts and competitors. If those are not in your files, the agent should not make them up, but you should still check every name, number and claim before sharing the document outside your organisation.
- No scripts are run by the skill
- No network use
- Review named customers, figures and competitors against your source files before any external use
Tips for better summaries
Fill in validation.md and background.md before you run the skill. The Problem, Who It's For and Why Now sections depend on that evidence. Keep mockups in supporting/mockup-ascii.md so the Visual Preview section has screens to pick from.
Use the skill only for features that are ready for development. The skill's structure assumes that stage, and it is not designed for public announcements where unverified claims would be risky.
- Write SPEC.md and validation.md first; the summary is only as grounded as they are
- Keep your folder names and file names matching the expected layout
- Read the Key Decisions section closely; it is meant to highlight surprising choices
What people use it for
- Giving sales a one-page pitch for a feature that is ready to build, drawn from the same spec engineering uses
- Briefing leadership on the problem, the why-now case and the success metrics in one document
- Handing engineering a behavioural overview with key decisions, dependencies and what is deferred
- Keeping one shared summary per feature folder that stays in the same structure across the team
Frequently asked questions
How do I install the exec-summary agent skill?
Download the ZIP from the page where you found this guide and unzip it into ~/.claude/skills/ for personal use or into a project's .claude/skills/ folder. In the Claude apps, you can instead upload it under Settings → Capabilities → Skills.
What does the exec-summary skill require?
It needs a feature folder containing SPEC.md, background.md, validation.md and supporting/mockup-ascii.md. It also needs an agent runtime that can read and write files in the project.
Is the exec-summary skill free to use?
The source lists the licence as Apache-2.0, which is the licence of the mvschwarz/openrig repository the skill belongs to. Check the licence file in that repository for the full terms. The source does not state any price.
Does the exec-summary skill run scripts or use the internet?
No. According to its listing, it runs no scripts and uses no network. It reads and writes files in the feature folder.
What if the summary has made-up customers or numbers?
The skill is designed not to invent them, but it can do so when the feature has no spec or validation files to draw on. Check the feature folder has real inputs, and verify every name, dollar amount and competitor before sharing externally.
Is the exec-summary skill a good alternative to writing summaries by hand?
It suits teams whose feature folders already follow the expected layout and who want one shared document for sales, leadership and engineering. The source does not compare it with other tools, so test it on one finished feature before you rely on it.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Use this skill if you are a product manager with ready-to-build features whose folders already hold SPEC.md, background.md, validation.md and mockups, and you want one standard summary for three audiences. Pick something else if your features have no spec or validation evidence yet, your team uses a different folder structure, or you need a public announcement where every claim must be verified.
Download the skill