# AI use cases for Open (limelit-co/open) Repo: [limelit-co/open](https://github.com/limelit-co/open) | Use cases: 4 | Page: https://dedupedai.com/use-cases/limelit-co-open/ Limelit Open is a self-hosted AI visibility tracker, shipped as one Go binary with SQLite, that measures how ChatGPT, Claude, Perplexity, Gemini and Google AI surfaces mention and cite your brand against competitors. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Track your brand's AI answer visibility against rivals: https://dedupedai.com/use-cases/limelit-co-open/track-brand-ai-visibility-vs-competitors/ - Let Claude answer AI visibility questions from your data: https://dedupedai.com/use-cases/limelit-co-open/ask-claude-about-ai-visibility-data/ - Find which sites AI engines cite for your category: https://dedupedai.com/use-cases/limelit-co-open/find-sources-ai-engines-cite/ - Export all visibility data as JSON or CSV: https://dedupedai.com/use-cases/limelit-co-open/export-visibility-data-for-reporting/ ## How to track your brand's AI answer visibility against rivals with Open Repo: [limelit-co/open](https://github.com/limelit-co/open) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Run one self-hosted pass that shows how often AI engines mention your brand versus competitors. **Who it is for:** A marketing or SEO lead who needs a defensible number for how often ChatGPT, Claude, Perplexity and Gemini mention their brand, and who does not want a closed SaaS holding the prompts and data. **The problem:** AI assistants now shape which brands buyers hear about, but there is no trustworthy, auditable view of whether your brand appears in their answers or how you compare with competitors. **The outcome:** A local dashboard with visibility, share of voice and a competitor ranking, where every number shows the sample of answers it rests on. ### Steps 1. Run `mkdir -p ~/limelit && cd ~/limelit && curl -fsSL https://github.com/limelit-co/open/releases/latest/download/limelit_$(uname -s | tr A-Z a-z)_$(uname -m | sed 's/x86_64/amd64/;s/aarch64/arm64/').tar.gz | tar xz limelit` to download the binary. (inferred, check the README) 2. Run `./limelit serve` and Open localhost:1515 in a browser. 3. In the setup wizard, enter your domain, your category, up to five competitors, and one key (the free Limelit Cloud allowance via `limelit login`, or your own OpenAI key). 4. Press Run, then read the Overview as each engine returns answers; numbers with fewer than 20 answers are marked as thin. ### Prompt to paste into your agent ```text Using the limelit-co/open repository, download the binary, start `limelit serve`, and complete the setup wizard for the domain with competitors . Press Run and report the visibility and share of voice per engine. Verify that each figure shows its answer count n. ``` ### Caveat Results come from vendor APIs with web search on and from scrapers for Google AI Overview, Google AI Mode and Bing Copilot, so they are not the same as what a person sees in the consumer apps. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/limelit-co-open/track-brand-ai-visibility-vs-competitors/ ## How to let Claude answer AI visibility questions from your data with Open Repo: [limelit-co/open](https://github.com/limelit-co/open) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Connect Claude to a local Limelit instance so it can read your visibility data in plain language. **Who it is for:** A marketer or founder who already uses Claude and wants to ask questions like which competitor is ahead on a given engine without building dashboards. **The problem:** Visibility data sits in a dashboard that someone has to read and interpret, which slows down questions from the team. **The outcome:** Claude answers questions about your brand's AI visibility by calling the Limelit MCP tools against your own SQLite data. ### Steps 1. Complete the install and first run from the quick start so that `~/limelit` contains the `limelit` binary. 2. Run `claude mcp add limelit -s user -- sh -c 'cd ~/limelit && exec ./limelit mcp'` to register the server with Claude Code. 3. For Claude Desktop, add the `limelit` entry with command `sh` and args `-c` and `cd ~/limelit && exec ./limelit mcp` to `claude_desktop_config.json`. 4. Ask Claude for your current visibility; the first tool call reports whether the instance is not set up, never run, or ready. ### Prompt to paste into your agent ```text In the limelit-co/open repository, register the MCP server with `claude mcp add limelit -s user -- sh -c 'cd ~/limelit && exec ./limelit mcp'`. Then ask it for the visibility of against across all engines. Expected output is a plain-language summary with the sample size behind each figure. Verify that the first tool call reports the instance state. ``` ### Caveat The MCP server reads an instance that must already be set up and have run at least once, so a fresh install will report that it has not been run yet. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/limelit-co-open/ask-claude-about-ai-visibility-data/ ## How to find which sites AI engines cite for your category with Open Repo: [limelit-co/open](https://github.com/limelit-co/open) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README See which pages engines trust, classified as your own, competitor, social, informational or other. **Who it is for:** A content or PR lead who needs to know which publications and pages AI engines cite before deciding where to earn coverage or update their own pages. **The problem:** Teams guess which sources influence AI answers, so they spend effort on pages that engines never use. **The outcome:** A ranked list of cited sites and a citation split by source type, showing which pages to appear on or displace. ### Steps 1. Run `./limelit serve` in the folder that holds the binary and Open localhost:1515. 2. Complete the setup wizard and press Run so that answers and their citations are stored. 3. Open the citation analysis view to read the citations split by source type and the top cited sites. 4. Open the Competitors page, which suggests the sites your answers cite most. ### Prompt to paste into your agent ```text Using limelit-co/open, run one pass for the brand in the category . Export the citation data and list the top 10 cited sites with their source type. Verify that each classification matches the stored citation rows. ``` ### Caveat Query fan-out shows the searches an engine ran, which may differ from the question you asked, so citation patterns reflect those searches as well as your prompts. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/limelit-co-open/find-sources-ai-engines-cite/ ## How to export all visibility data as JSON or CSV with Open Repo: [limelit-co/open](https://github.com/limelit-co/open) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Pull the whole instance into one JSON file or a folder of CSVs for your own analysis or board material. **Who it is for:** An analyst who must put AI visibility figures into a spreadsheet, BI tool or board deck and needs a reproducible copy of the underlying rows. **The problem:** A metric that goes into a board deck should be recomputable from stored data, but closed tools rarely export it. **The outcome:** A complete export of prompts, answers and metrics that can be recomputed outside the tool, optionally limited to recent rows with `--since`. ### Steps 1. Stop at a working instance that has completed at least one run, in the folder that holds the binary. 2. Run `limelit export` to write the whole instance as one JSON document or a directory of CSVs. 3. Use the `--since` filter to limit the export to rows from a chosen date onward. 4. Load the output into your spreadsheet or BI tool and recompute one figure from the formulas in the README to confirm it matches the dashboard. ### Prompt to paste into your agent ```text In the limelit-co/open repository, run `limelit export` for the instance in ~/limelit and write the output to . Expected output is a JSON document or CSV directory covering all answers. Verify that the row count for matches the dashboard count for that engine. ``` ### Caveat The export contains your prompts and every stored answer, so keep the output file as private as the SQLite database it came from. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/limelit-co-open/export-visibility-data-for-reporting/