← ClaudeAtlas

visibility-progress-reportlisted

Create an executive GEO visibility progress report from time-series LLM prompt execution data. Use when the user needs a monthly or quarterly AI-search visibility update, progress narrative, performance report, platform gap analysis, topic trend summary, or executive briefing from prompt tracking exports.
birdseyeglobal/portage · ★ 0 · Data & Documents · score 68
Install: claude install-skill birdseyeglobal/portage
# GEO Visibility Progress Report Build a progress narrative from prompt execution data. This skill is for recurring reporting after a brand has enough tracking history to show movement. If there is no time series, produce a baseline visibility report instead and label it as such. ## Inputs Required: - Prompt execution CSV. - Brand name. Useful columns: - `Execution Date` - `Prompt` - `Topic` - `Platform` - `Workspace Brand Mentioned` - `Competitor Mentions` - `Citations (Links)` - `AI Response` Optional: - Published content count. - Draft pipeline count. - Owned URL list. - Article title mapping. - Business context for the reporting period. ## Workflow 1. Parse and normalize dates, platforms, topics, brand mentions, competitors, and citations. 2. Check whether the data has enough history for trend reporting. Default threshold: at least 4 weeks. 3. Compute: - Current visibility rate. - First-period versus latest-period delta. - 7-day or period moving average when dates support it. - Topic-level visibility. - Platform-level visibility. - Owned URL citation counts. - Competitor mention context. 4. Identify what changed and why. 5. Write an executive summary and visual/report outline. 6. Recommend next focus areas. ## Helper Script Use `scripts/build_report.py` when the user wants a PDF progress report from a prompt execution CSV: ```bash python3 <skill-path>/scripts/build_report.py \ --csv path/to/prompt-executions.csv \ --brand "Brand Name