agentleFS
awaikawaik/direct-mcp-ai-project/.claude/skills/demand-research/SKILL.md

demand-research

Skill20 starsChanged 4 days agoMITView on GitHub

Исследовать поисковый спрос через LidFly Wordstat: сезонность, упущенный спрос, интент, регионы и каннибализация. Использовать для проверки спроса и подготовки рекомендаций без client_login или connection_id Директа.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 3 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use for demand checks, seasonality, missed demand, keyword cannibalization, region demand, and intent verification.

    line 8
  2. Very specific:

    Use wordstat_* through LidFly MCP v3.

    line 12
  3. Very specific:

    Do not pass client_login, connection_id, or advertising account scope.

    line 13
  4. Very specific:

    Use wordstat_find_region when regional ids are needed.

    line 22
  5. Very specific:

    Use wordstat_top_requests, wordstat_dynamics, and wordstat_regions as needed.

    line 23
  6. Distinctive:

    If needed, read the unfinished month separately with daily/weekly aggregation.

    line 35
  7. Distinctive:

    Keep annual dynamics separate from the last-30-days regional distribution.

    line 36

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

Demand Research

Use for demand checks, seasonality, missed demand, keyword cannibalization, region demand, and intent verification.

Wordstat Rules

  • Use wordstat_* through LidFly MCP v3.
  • Do not pass client_login, connection_id, or advertising account scope.
  • Get schemas before first call.
  • wordstat_top_requests is the primary exact frequency source for the last 30 days.
  • check_search_volume only checks whether Direct has traffic; it is not exact Wordstat frequency.

Workflow

  1. Clarify product, geography, audience, and business goal.
  2. Collect seed queries and stop topics.
  3. Use wordstat_find_region when regional ids are needed.
  4. Use wordstat_top_requests, wordstat_dynamics, and wordstat_regions as needed.
  5. Classify intent: commercial, informational, comparison, branded, competitor, support.
  6. Identify missed demand, seasonality, and cannibalization risk.
  7. Return the research in the requested format. Save to Workspace only when the user asked to save there and the exact project scope is resolved; usefulness for future work is not permission to persist it.

For a business topic rather than explicitly requested exact phrases, decompose the topic into 3–5 short seed queries (usually 1–4 words). Measure each separately; do not send the whole business description as one Wordstat phrase or add overlapping phrase counts together. Preserve explicitly requested phrases without substitution.

For “the last 12 months”, use the last 12 complete calendar months with monthly dynamics, ending at the end of the previous month, and state the actual dates. If needed, read the unfinished month separately with daily/weekly aggregation. Keep annual dynamics separate from the last-30-days regional distribution. Country, district, region and city rows are nested: use parent IDs/names and never sum counts or shares across those levels. For annual trends in named regions, request dynamics with those region filters. Missing dates are unconfirmed, not measured zero.

Output

Bind every number to its exact phrase, section, regions, devices and observation time. Use typed observations in embedded chat. Top/regions cover the last 30 days; do not invent calendar boundaries. Dynamics uses the actual normalized interval. Show unknown/failed values as N/A, preserve measured zero, and never turn YES/NO into frequency. Separate unmeasured ideas from measured phrases. Report unread sites and missing sources, and never infer conversion quality from frequency alone. Advice based on supplied data does not automatically require paid Wordstat reads.

Give clusters with intent, frequency notes, negative themes, recommended landing pages/campaigns, and the actual saved artifact only if a save was requested and verified.

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.