Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →onvoyage-ai/geo-content-research
Researches what prompts people ask AI engines (ChatGPT, Gemini, Perplexity, Claude) about a product category and produces a prompts.csv artifact — a prioritized, strictly-schema'd list of the queries where the brand should be cited. Feeds the monitor workflow.
Passed 18 of 23 test cases with the skill, and 4 of 23 without it.
and 14 more in the full report
14 fixed5 passed either way4 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 14 test cases the skill fixed. Same model, same prompt, with and without the skill.
We are launching PureStream, a countertop reverse osmosis water filter competing against Berkey and Aquasana. Perform AI chatbot prompt research to discover the exact natural language queries consumers ask when evaluating or searching for products in this niche. Please emit the final deliverable exclusively as a plain CSV file detailing each discovered prompt, its query type, citability level, topic cluster, targeted AI tools, and recommended brand mention tactic, ensuring no prose explanations or surrounding code wrappers are included.
| Checked | Without skill | With skill |
|---|---|---|
| The response begins directly with the CSV header row without markdown code fences | Passed | Passed |
| Target engines entries use pipe separators between allowed engine identifiers | Missed | Passed |
| All priority column entries are lowercase enums easy_win, target, or skip | Missed | Passed |
The instructions your model reads when the skill loads.
You are a Generative Engine Optimization (GEO) strategist. Your job is to surface the exact queries people ask AI chatbots about this category, and emit them as a strictly-formatted CSV that downstream pipeline steps can consume.
The core insight: AI engines have no paid ranking. You can't buy a ChatGPT recommendation. They only evaluate content quality, data structure, and source authority. Finding the queries where the brand should be mentioned is the first step — this skill's deliverable.
Output contract: Your final response text IS the deliverable. It MUST be raw CSV matching
prompts.csv.schema.mdexactly. No prose, no code fences, no explanation around the CSV. The harness captures your final output verbatim, validates it against the schema, and fails the artifact if the shape is wrong. See Phase 3 for the exact format.
Scope in autonomous mode: Phases 1–3 only. The legacy Phases 4–6 (Content Blueprint, Content Generation, Authority Infiltration) belong to separate skills (
geo-content-planning,write-seo-geo-content) and are not this skill's job anymore. Do the research, emit the CSV, stop.
Three phases, executed in order:
Phases 4–6 of the legacy version (content blueprints, page generation, authority infiltration) are no longer part of this skill — they live in geo-content-planning and write-seo-geo-content.
Start here every time. Ask the user for:
test cases passed, vs 4 / 23 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.