Builds or refines an AI visibility tracking strategy: the prompt set, grouping, brand roster, and reporting split for ChatGPT, Gemini, and similar.
ai-visibility-tracking-strategy-builder
Build or refine an AI visibility tracking strategy — the prompt set, grouping structure, brand roster and reporting split that measure how a brand appears in ChatGPT, Gemini, Perplexity, AI Overviews and similar — on any AI visibility monitoring platform. Runs as an iterative loop (Intake → Strategy → Write → Analyse) with an optional stakeholder presentation. Load whenever the user mentions AI visibility, GEO (generative engine optimisation), AI brand monitoring, prompt tracking, "what prompts should we track", "how are we showing up in ChatGPT / Gemini / Perplexity", AI mention tracking, a prompt set for an AI monitoring tool, or a tracking strategy review. This is the platform-agnostic core of a skill family: load it together with the companion for the platform in use (Peec AI: `peec-ai-tracking-strategy-builder`; SISTRIX: `sistrix-tracking-strategy-builder`; searchVIU: `searchviu-tracking-strategy-builder`; LLM Pulse: `llmpulse-tracking-strategy-builder`), or alone for any other tool.
معلومات المصدر
- المستودع
- rebelytics/ai-visibility-tracking-strategy-builder
- آخر نشاط في المصدر
- ١٦ نوفمبر ٢٠٢٦ في ١٣:١٣
- لغة SKILL.md المكتشفة
- الإنجليزية
- النجوم
- ٣
- التفرعات
- ١
AI Visibility Tracking Strategy Builder
Builds or refines an AI visibility tracking strategy: the prompt set, grouping, brand roster, and reporting split for ChatGPT, Gemini, and similar.
Uses
Use this Skill to build or refine an AI visibility tracking strategy for a brand — the prompt set, grouping structure, brand roster, and reporting split — on whatever AI visibility platform the brand uses. It covers new projects with no prompts yet and existing projects that need rationalising. It is not a platform reference, not a content-strategy Skill, and not an ongoing reporting Skill beyond what each Analyse step needs for the next iteration.
Prerequisites
The brand needs access to an AI visibility monitoring platform. This Skill is the platform-agnostic core; when a published companion exists for Peec AI, SISTRIX, searchVIU, or LLM Pulse, the pinned SKILL.md says to load that companion alongside it, and those companions were not drafted here. Platform field names, API or MCP mechanics, plan-tier limits, and import formats live in the companion or in the platform's own tool skill, which this draft did not read or execute. Detailed Intake, Strategy, Write, Analyse, Phase B, and quality-gate playbooks live in unattached references/ files and were not treated as pinned capabilities. The site's general install module covers Skill-file installation. This draft did not connect to any monitoring platform, run references/allocate.py, or execute the Skill.
How to use
Tell the agent which brand and platform to work with, and share any brand context or data sources already on hand. On the first loop the agent should say that iterations usually run about a day apart, that trend-level questions may need 7 to 14 days, and that a stakeholder presentation is optional later; it must not start Analyse in under 24 hours. Phase A then loops through intake, a prescriptive strategy the user signs off, writes to the platform (or a validated import file where that is the contract), and analysis that feeds the next loop. User-provided data is asked only as a batched fallback after automated sources. Once the strategy has stabilised, the agent may offer an optional Phase B stakeholder presentation from existing artefacts. Working artefacts such as intake state and a verification log must exist, but their format is the user's choice. This is a source-documented process overview; the Skill was not executed in this draft.
Limitations
This Skill writes the configured tracking project through the platform API after user sign-off, or delivers a validated import file where that is the platform's contract; it does not execute content or distribution work it may surface. Branded and non-branded prompts measure different things and must not be rolled into one visibility score. Do not default to tracking every engine; coverage should match where the audience actually asks, and plan tiers may limit the engine set. Fresh prompts need time: the documented default iteration pace is daily, and Analyse should not run in under 24 hours; share-of-voice or sentiment trends are documented as more stable after 7 to 14 days. Specific review cadences such as 30 or 90 days are described as user-configurable defaults, not skill-validated prescriptions. Companion skills for Peec AI, SISTRIX, and searchVIU, and the observer skill named in the contributing section, were not drafted here. Detailed phase rules in unattached references/ were not treated as pinned capabilities. The Skill says it is not affiliated with any AI visibility platform and that no vendor has reviewed or endorsed it. The original reply followed the reviewed 1.7.0 source. This explanation is based on the later 1.9.0 revision, which also lists an LLM Pulse companion. The parent post is context. The LICENSE and SKILL.md both state CC BY 4.0. This draft did not run the Skill or write to any platform. The agent must check what it can already access before asking the user for anything. The platform's own API or MCP, connected data MCPs (analytics, search console, SEO tools, commerce platforms, crawl tools), other loaded skills (brand dossiers, business context), and earlier conversation context all take precedence. Any data the user provides manually — domain lists, market priorities, taxonomies, customer-voice samples, regulatory notes — is saved to the persistence store so the user never has to provide it twice. Any CSVs, exports, or documents the user supplied during Ring 3. Kept so future loops don't re-ask. In environments without persistent storage, the working artefacts live inside handoff docs at session end and are re-seeded at session start.
خيارات التثبيت
يُحدَّد Prompt الذي يراجع المصدر أولًا بشكل افتراضي. يمكنك التبديل إلى أمر مباشر أو تنزيل نسخة محلية.
مراجعة ملفات المصدر
اقرأ SKILL.md وأي ملفات مرافقة يعرضها SkillsMP قبل أن تقرر التثبيت.
عرض SKILL.md
| name | ai-visibility-tracking-strategy-builder |
|---|---|
| description | Build or refine an AI visibility tracking strategy — the prompt set, grouping structure, brand roster and reporting split that measure how a brand appears in ChatGPT, Gemini, Perplexity, AI Overviews and similar — on any AI visibility monitoring platform. Runs as an iterative loop (Intake → Strategy → Write → Analyse) with an optional stakeholder presentation. Load whenever the user mentions AI visibility, GEO (generative engine optimisation), AI brand monitoring, prompt tracking, "what prompts should we track", "how are we showing up in ChatGPT / Gemini / Perplexity", AI mention tracking, a prompt set for an AI monitoring tool, or a tracking strategy review. This is the platform-agnostic core of a skill family: load it together with the companion for the platform in use (Peec AI: `peec-ai-tracking-strategy-builder`; SISTRIX: `sistrix-tracking-strategy-builder`; searchVIU: `searchviu-tracking-strategy-builder`; LLM Pulse: `llmpulse-tracking-strategy-builder`), or alone for any other tool. |