agentleFS
SpytrendSpytrend/spytrend-mcp/skills/spytrend-webmaster-overview/SKILL.md

spytrend-webmaster-overview

Skill30 starsChanged 3 months agoMITView on GitHub

Profile the advertising network behind a domain or a group of ads with SpyTrend MCP data: connected domains, advertiser pages, identifiers tying ads to one operator, target countries and the creatives that operation reuses most, from a database of 1B+ tracked Meta ads. Use when the user wants to know who is behind a set of ads, see everything one operator or competitor runs across pages and domains, or check whether identical ads on different pages belong to one player. Triggers: "ads spy across pages", "who is behind these ads", "everything this operator runs", "the whole operation", "network behind a landing domain"; RU «кто стоит за рекламой этого домена», «какие ещё домены у этого рекламодателя», «вся связка страниц оператора». Not for ranking advertiser pages in a market (use spytrend-advertiser-overview) and not for a website's traffic review (use spytrend-website-review).

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 1 is distinctive, and 8 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 8

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read references/spytrend-contract.md before making any data call.

    line 20
  2. Very specific:

    Call get_usage and disclose the planned depth and media choice.

    line 26
  3. Very specific:

    If the input is not already a canonical id, call search_webmasters by exact domain or query with a small limit.

    line 27
  4. Very specific:

    Read market size from pagination.total only with its status.

    line 30
  5. Very specific:

    Call get_webmaster with the chosen canonical id.

    line 32
  6. Very specific:

    Call search_creatives with webmaster_id, optional category and sort_by=total_ads for the exact global network slice.

    line 43
  7. Very specific:

    Call search_ads with webmaster_id, the selected categories and countries, country_match=any, status_today=active, sort_by=date, dedupe=true …

    line 47
  8. Distinctive:

    If the card contains a network-specific creative slice, use its slice order for "most reused by this network".

    line 52
  9. Very specific:

    Use references/report-template.md or the packaged renderer with references/report-template.html.

    line 65

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

SpyTrend advertising network overview

Read references/spytrend-contract.md before making any data call. The skill name preserves the MCP entity name for routing; user-facing output says "advertising network" or "network".

Resolve one network

  1. Call get_usage and disclose the planned depth and media choice.
  2. If the input is not already a canonical id, call search_webmasters by exact domain or query with a small limit. Show ambiguous candidates and ask the user to choose rather than merging them.
  3. Read market size from pagination.total only with its status. Never read a top-level total and never promise a complete list from a capped response.
  4. Call get_webmaster with the chosen canonical id. If analytics_pending=true, treat counters as pending, not zero.

Bounded profile

  1. Render the returned domains, advertiser pages, the identifiers that tie ads to one operator, and countries with their returned counts, caps and statuses. A missing inventory is unavailable, not proof that no identifier exists.
  2. Present attribution edges as evidence of observed combinations. A small edge does not prove common ownership, and inherited identifiers do not describe current operation by themselves.
  3. Call search_creatives with webmaster_id, optional category and sort_by=total_ads for the exact global network slice. Do not add country: the current read path cannot preserve the exact network slice together with that filter. Use returned geo fields as context, not as a country ranking.
  4. Call search_ads with webmaster_id, the selected categories and countries, country_match=any, status_today=active, sort_by=date, dedupe=true and a small explicit limit for fresh examples. This means the selected country is present in the ad's known geographies; it does not mean that country is dominant. Label dates according to the returned date axis.
  5. If the card contains a network-specific creative slice, use its slice order for "most reused by this network". Keep market-wide family size as separate context; do not use it to reorder the network slice.
  6. With confirmation, retrieve media for only the chosen rows. A representative ad is a cheaper placement example; a creative fetch represents the material family. State which path was used.

Report

Include resolution evidence, current/pending status, returned identifier coverage, country activity, creative rankings, fresh examples, period labels and spend. Say "processed N of M" whenever a bounded cap is used.

Use references/report-template.md or the packaged renderer with references/report-template.html; otherwise return the same report in chat.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.