canonical: https://jentic.com/apis/aimentiontracker.ai/aimentiontracker # AIMentionTracker API The AIMentionTracker API reports how often AI assistants name and cite a brand and which sites they read instead. It returns presence per engine and per brand over a trailing window, with each engine reported separately rather than averaged, plus the stored answers behind those numbers with their mentions and citations. A sources view lists the domains the engines read before answering, and an alerts feed surfaces findings newest-first when a brand's visibility moves beyond its margin of error. GET /me returns the live thresholds and the rules for reading the numbers. ## For AI agents Read how AI assistants mention and cite a brand: presence per engine, the answers behind the numbers, the sources engines read, and alerts, for AI agents. ## Scope Does not post content, run ad campaigns, or change what AI assistants say. Use for reading how AI assistants mention and cite your brand only. ## Capabilities - Read a brand's presence per AI engine over a trailing window, reported separately per engine - Retrieve the stored AI answers with their brand mentions and citations - List the source domains the engines read before answering - Review findings about a brand from the alerts feed, newest first - List the brands a key may read and the rules for reading the numbers ## Use cases ### AI visibility reporting agent An AI agent building a weekly brand report needs presence figures that respect how the data must be read. Through Jentic the agent searches for 'read brand visibility per engine', loads the AIMentionTracker operation with its input schema, and reads visibility with the key injected at call time. Because each engine is reported separately rather than averaged, the agent presents per-engine figures instead of a single blended number. Example prompt: Read GET /visibility for a brand over a trailing window and report presence per engine without averaging across engines ### Citation and source analysis A content team wants to know which sites AI assistants read instead of their own. AIMentionTracker returns the stored answers with their citations and a sources view of the domains engines consult before answering. An agent can pull the sources, rank them by frequency, and flag competitor domains that are being cited. Example prompt: Read GET /sources for a brand and GET /answers to list the domains cited most often in AI answers ### Visibility alerting A marketer wants to hear only about real moves in brand visibility, not noise. AIMentionTracker suppresses findings that sit inside their own margin of error and surfaces the rest through an alerts feed, newest first. An agent can poll the alerts, read the thresholds from the account endpoint, and summarise only the findings that cleared the margin. Example prompt: Poll GET /alerts for a brand and summarise each finding, using GET /me to report the thresholds applied ## Key endpoints | Method | Path | Description | | --- | --- | --- | | GET | `/visibility` | Presence per engine and brand over a trailing window | | GET | `/answers` | Stored answers with mentions and citations | | GET | `/sources` | Domains the engines read before answering | | GET | `/alerts` | Findings about a brand, newest first | | GET | `/brands` | Brands this key may read | | GET | `/me` | Key identity, thresholds, and reading rules | ## Key resources - **Visibility** — Presence per engine and per brand over a trailing window - **Answers and sources** — Stored answers with mentions and citations, and the domains engines read - **Alerts and account** — Findings feed and the key's brands, thresholds, and reading rules ## AI readiness This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use. - **Score:** 49 / 100 - **Maturity:** Foundational - **Dimensions:** - Foundational Compliance: 93 / 100 - Developer Experience & Jentic Compatibility: 63 / 100 - AI-Readiness & Agent Experience: 26 / 100 - Agent Usability: 94 / 100 - Security: 50 / 100 - AI Discoverability: 50 / 100 - **View full report:** https://jentic.com/apis/aimentiontracker.ai/aimentiontracker/scorecard - **How the score is calculated:** https://docs.jentic.com/reference/api-readiness-framework/overview/ - **More about the dimensions:** https://docs.jentic.com/reference/api-readiness-framework/specification/#dimensional-model-overview ### Score it yourself Every API in the directory is allowlisted, so you can re-score it with no key required. - **Score your own API:** https://jentic.com/scorecard.md - **Scoring CLI agent skill:** https://github.com/jentic/jentic-api-scorecard/blob/main/skills/jentic-api-scorecard/SKILL.md ```sh npx @jentic/api-scorecard-cli score ``` ## Why Jentic - **Setup:** Wiring AIMentionTracker by hand means obtaining an organisation key, sending it as an X-API-Key or bearer header, and carrying the reading rules so per-engine numbers are not averaged. Through Jentic you install once, import AIMentionTracker from the API Directory, store the key once, and your agent calls it. - **Permission scoping:** The API is read-only with brands scoped to the key, and each view sits at its own path such as /visibility and /alerts, so a rule can pin your agent to the reads it needs. You choose which operations it may call, so the answers or sources endpoints are not included unless you add them. - **Credential handling:** Your AIMentionTracker key is stored once, encrypted, by your own Jentic One instance and attached as the X-API-Key or bearer header at execution time. It never enters the agent's prompt, logs, or context. - **Discovery method:** Agents search Jentic by intent such as 'read brand visibility per engine' or 'check brand alerts', and Jentic returns the matching AIMentionTracker operation with its input schema so the agent calls the right endpoint over the trailing window it needs. ## Related APIs - **SEMrush API** — Keyword, rank, and SEO visibility data for traditional search engines - **serpstack API** — Live search engine results pages for comparing classic SERP presence to AI mentions ## FAQ ### What authentication does the AIMentionTracker API use? It authenticates with an API key sent in the X-API-Key request header, and the spec notes an Authorization bearer header is accepted equivalently. Keys belong to the organisation rather than a person, so they survive staff changes. Through Jentic the key is held by your own instance and attached per call. ### How should an agent read the visibility numbers from the AIMentionTracker API? Each engine is reported separately and should not be averaged, since engines disagree and a blend describes no real surface. A null rate means there were not enough answers yet rather than zero, and an empty alerts list does not mean nothing changed, because moves inside the margin of error are suppressed. GET /me returns the live thresholds behind these rules. ### What are the rate limits for the AIMentionTracker API? The OpenAPI spec does not declare numeric rate limits. Check the AIMentionTracker documentation for the current per-plan allowance, and note that most endpoints return a trailing window, so frequent polling rarely adds new data. ### Is there an AIMentionTracker MCP server? You do not need an MCP server to give your agent AIMentionTracker. Jentic connects it directly from the API Directory: import it, store your key once, and your agent calls the visibility, answers, and alerts operations it needs. ### How do I read brand visibility through Jentic? Search Jentic for 'read brand visibility per engine', load the GET /visibility operation, and run it for a brand over a trailing window. The key is attached by your instance, and the response keeps each engine separate. To run it on your own infrastructure, install Jentic One from its GitHub repo. ### Can I limit what my agent is allowed to do with the AIMentionTracker API? Yes. The API is read-only and a key may only read its own organisation's brands, so a rule can allow the agent just the reads a task needs, such as visibility and alerts, while leaving the answers or sources endpoints out unless you add them. The agent stays bound to the operations you grant, and every call is logged.