Tools
PromptWatch
How your brand shows up in AI answers: prompts, monitors, citations and sentiment.
Official site: promptwatch.com
What it does
- Track a set of prompts in a monitor across ChatGPT, Claude and Gemini
- Read visibility over time, and compare against competitors on a heatmap
- Read the answers themselves, their sentiment and the sources they cite
The citations are the half that can be acted on. A model's answer names where it got the claim, which turns "we are invisible in AI answers" from a ranking complaint into a specific list of pages that do not say what someone would need to read. Visibility moves when those sources move.
Why through Tulina
Most systems now publish their own way for a model to reach them, and that is a good thing — it is what Tulina connects to. The connection was never the hard part. Deciding who may use it, what it is allowed to return, and what the model knows before it asks: that is.
The keys stay in one place
Connect PromptWatch once. The credential is encrypted in a vault, never held in clear, and never pasted into a chat or a workflow file. A person uses their own key where they have one, their team's where they do not, the organisation's after that — so nobody is blocked and nobody is handed more access than they need. Revoke it once and everything running on it stops.
What leaves is filtered before it leaves
Roles decide who can reach PromptWatch at all. Beyond that, a policy on the connector itself decides what comes back: fields dropped, identifiers masked or hashed, people pseudonymised. The data is minimised at the source rather than after it has already reached a model.
Every call is on the record
Runs are logged for the organisation — what was called, by whom, when. That is what makes an automated action reviewable after the fact, and what a security review asks for first.
The run starts knowing your company
A direct connection gives a model access. It does not tell it how your company names things, which accounts matter, or what was decided last quarter. Tulina carries that across every tool, so the same request against PromptWatch returns something usable instead of something generic.
Processes that run on PromptWatch
None published yet.