Ad Library research

Read-only, public data. Read meta-ads-core first.

Preflight: two checks, kept separate

Independent questions; never let one answer stand in for the other.

1. Meta's official Ads MCP - the execution layer. Look for tools named ads_* in this session. With none, nothing here can reach an account: say so and help the user connect it - connect-meta-mcp.md under references/ in the meta-ads-core skill, or https://github.com/bartlomiejborzucki/meta-ads-agent/blob/master/skills/meta-ads-core/references/connect-meta-mcp.md. The local CLI does not replace it.

2. The local meta-ads-agent CLI - optional, usually absent. Assume it is missing until this succeeds:

meta-ads-agent --version

"command not found" is the normal answer, not a fault. Show no meta-ads-agent ... command before that probe has succeeded - a command that fails at the user's prompt costs more than the step it saves. Say "that step needs the optional CLI, which is not installed here" and carry on.

The MCP alone covers audits, reporting, Ad Library research, previews, tracking, creative, optimisation and the changes that follow; .meta-ads/ files can be written directly from the templates in meta-ads-core's assets/. Only these need the CLI: plan validation (so campaign builds), local image and video upload, video / existing-post / multi-variant / carousel creatives, deletion, and the report arithmetic (which has a by-hand route).

ads_library_search   → search the public Meta Ad Library

What this is for

Understanding what a market is already saying, so the creative you write is positioned against it rather than accidentally identical to it.

Extract themes. Do not clone.

Two reasons the distinction matters. Copying creative is a legal problem - someone else's ad is their copyrighted work. And ads that sound like the category perform like the category: the gap is usually where the opportunity is, not the crowd.

A competitor's claim is theirs to substantiate, not yours to borrow. If a competitor says "cut costs 40%", you may not say it. You have no evidence for it, and it is their number.

Reading the results

Longevity is the most useful signal available

The Ad Library shows how long an ad has been running. A long-running ad is the closest thing to a public performance signal you can get, because advertisers do not keep paying for ads that lose money.

Caveats that keep it honest: a large advertiser may run an ad for brand reasons regardless of performance, a small advertiser may simply not be watching, and you cannot see spend. So it is evidence, not proof.

Volume is not quality

Twenty variants of one ad means the advertiser is testing, not that the ad works. Judge concepts, not counts.

You cannot see performance

No spend, no CTR, no conversions, no cost per result. Anyone claiming to know which competitor ad is "winning" from the Ad Library is inferring from longevity. Say so when you do it.

A useful search pattern

1. Direct competitors by advertiser name. What they lead with, and how long the current set has run.
2. Category terms by keyword. Who else is in this auction, including advertisers the user has not thought of.
3. Adjacent categories solving the same problem differently. Often where the unclaimed angles are.
4. Geography. The Ad Library is region-scoped, and a market can look entirely different in another country.

Then look for the absence. If every ad in a category leads with price, the angle nobody is using is available.

How to report it

# Ad Library — project management tools, PL, 2026-09-16

## Angles in market
1. Price / value          — 6 of 11 advertisers lead with it
2. Integration breadth    — 3
3. Speed of setup         — 2
4. Team visibility        — 1

## Long-running (likely working)
- <Advertiser A> — running 94 days, "onboarding in an afternoon".
  Angle: speed of setup. Format: 30s screen recording, captions, no voiceover.
- <Advertiser B> — running 61 days, free-trial offer, static with product UI.

## Formats
Static product screenshots dominate (8 of 11). Two use screen recordings.
None uses talking-head or UGC-style video.

## Nobody is saying
- What it costs to keep doing this manually
- Anything about the person who currently owns the spreadsheet
- Anything aimed at the team rather than the buyer

## Read on this
INTERPRETATION  The category competes on features and price. The cost-of-
                status-quo angle is unclaimed here, and it needs no
                competitive claim to make — which means we can run it
                without evidence we do not have.
CAVEAT          Longevity is the only performance signal available. No spend,
                no CTR, no conversions. A 94-day ad is probably working, but
                a large advertiser may run one for brand reasons regardless.

Describe competitors' angles in your own words. Do not paste their copy into a report - the report is where "inspiration" quietly becomes duplication.

Handing off to creative

Pass meta-ads-creative:

• the angles in market, and how crowded each is
• the unclaimed angles
• format conventions, and where they are unanimous enough to break
• the offer structures in use

Do not pass competitor copy. The creative skill writes from the offer brief, the brand voice, and the gap - which is the useful output of this research.

Limits

• Public data only. Not all ads, not all regions, not all time.
• No performance data of any kind.
• Coverage varies by country and by advertiser category.
• An absent advertiser may be advertising and not indexed here.

Say what you searched, in which region, and on what date. Ad Library results change constantly, so an undated finding is not reproducible - and a research note nobody can re-check is a rumour.