Builds an ad brief for any platform. Researches the user's product, audience, market, and KPIs proactively, then asks the user to confirm and fill gaps. Saves the result as .agents/ad-brief.md for future sessions. Not for writing ad copy, choosing platforms, configuring campaigns, or analyzing results.
Scanned 9/4/2026
Install to Claude Code
npx -y skills add NeverSight/skills_feed --skill ad-brief --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Ad Brief?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/neversight-ad-brief)More formats (shields.io, HTML) on the badges page.
---
name: ad-brief
description: >
Builds an ad brief for any platform. Researches the user's product, audience, market,
and KPIs proactively, then asks the user to confirm and fill gaps. Saves the result
as .agents/ad-brief.md for future sessions. Not for writing ad copy, choosing platforms,
configuring campaigns, or analyzing results.
---
# Ad Brief
This process is the same regardless of ad platform. Run it before any campaign work.
## How to interact
- **Use `AskUserQuestion` / `ask_user_question`** when asking about experience level, goals, or decisions. Present 2-4 selectable options - avoid making them type long answers. For questions where multiple answers are valid (e.g. target audiences, customer types), allow selecting multiple or typing a custom answer. Never force picking one when the user may have several.
- **Ask for reference materials early.** Ask if the user has existing documents you should read: brand guidelines, ICP/persona descriptions, competitor lists, landing page URLs, or previous ad performance data. If they provide files, read them and incorporate into your recommendations.
## Check for existing brief
**First, check if `.agents/ad-brief.md` exists** in the user's project. If it does, read it and skip this process.
If not, build one using the template at `template.md`. Two phases:
## Phase 1: Research (you do the work)
1. Ask the user for their product name and URL (or detect from the project: package.json, README, etc.)
2. Research proactively: visit their website, infer the audience, look up competitors, figure out the offer
3. Fill in as much of the brief as you can from what you found
## Phase 2: Confirm and complete (user fills gaps)
1. Present your filled brief to the user: "Here's what I found. Correct anything that's wrong and fill in what's missing."
2. For any remaining gaps, ask **one at a time** using `AskUserQuestion` / `ask_user_question`
3. Save the result as `.agents/ad-brief.md` in the user's project root. This persists for future sessions.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!