For the complete documentation index, see llms.txt. Every page on this site is also served as Markdown: append `.md` to any URL, or send `Accept: text/markdown`.
Attensira Logo
Attensira
Attensira for Content Teams

Know which pages
the models actually cite.

For editors and content leads: Attensira reads the answers ChatGPT, Perplexity, Google AI, Claude and Gemini give to the questions you track, shows which of your URLs they linked, and opens a pull request with the rewrite it thinks would change that.

Get StartedBook A Demo
•Cancel anytime•Nothing merges without your approval
Workflow

Read the answer, find the gap, ship the edit

Read what the models answered

Each question you track is put to each tracked model on a schedule, at most once per model and country per day, and the answer text is stored exactly as returned.

✦
Pull requests
Add comparison table to /compareMerged
Answer “good for small teams?”Merged
Rewrite /docs/api overviewOpen
Split /integrations by platformNeeds you

Review the edit as a branch

Each fix is a pull request on a fresh branch, labelled `attensira`, naming the page and the finding. Attensira then tracks whether you merged or closed it, and what CI said.

OpenAI
Perplexity
Claude
Gemini

Measure it again, honestly

The next reading reports whether your citation rate moved, and only calls it a move if it clears a two-proportion z-test at 95%. Otherwise it says the change is not provable.

Read, find the gap, ship the edit

Track the questions, see how they are answered

You write the buyer questions and group them into topics. Each one runs against each tracked model on a schedule — at most once per model and country per calendar day — and comes back as a mention rate and a citation rate with the number of successful runs behind them. Today's numbers are always partial, and Attensira says so.

See how tracking works
Tracked prompts · last 7 days · examplemention rate
best project management tool for remote teams40%
n = 15Cited too
how do I manage a distributed team13%
n = 15Named, never linked
project management vs task management0%
n = 15Measured zero
agile tools for startups—
not measuredModel not tracked
5
AI answer surfaces read on a schedule
Verbatim
Every successful answer stored and re-readable
Per URL
Citations, share and average position
PR
How the rewrite reaches your site
Built for Content Teams

What an editor gets

Evidence you can read, counts you can check, and edits that arrive as reviewable diffs.

Read the answers, not a summary of them

Attensira stores the verbatim text of every successful run, so an editor can read the paragraph where a competitor got named and you did not. Detection re-runs over that stored history rather than starting a fresh chart.

See how tracking works

Work from citations, not a content grade

Every cited page shows its citations in the window, its share of your domain's citations, and its average position in the model's citation list. There is no page score to optimise — the numbers are counts you can check.

See page-level citations

Get the draft as a diff, in your voice

The agent writes against your brand profile and Skills, then opens a pull request naming the page, the finding, and why the edit answers it. You read the diff, edit it, merge it or close it.

See how the agent writes

Save the recurring passes as automations

An instruction plus a cadence, with a stored run history — a weekly re-read of the prompts you care about, or an emailed summary. Anything that needs a human decision lands in the Inbox instead of happening quietly.

See automations

Where editorial work goes next

The surfaces that turn an answer you lost into a diff on a branch.

Content optimization, and the diff it producesKnowledge and Skills: the brand profile the agent writes againstPassages that survive extractionAI visibility for information and mediaThe AI search glossary

See which of your pages
the models cite today.

Track the questions your buyers ask, connect the repository your site lives in, and read the first pull request the agent opens.

Get StartedBook A Demo
•Cancel anytime
FAQ

Frequently Asked Questions

Not on GitHub?
Pull requests are the only way Attensira writes to your site today. Talk to us about what your stack needs.
Book A Demo

Which of your URLs models actually cite. The Pages view lists each cited URL with its citations in the window, its share of your domain's citations, and its average position inside the model's citation list. A page that answers a tracked prompt and is never cited is the gap worth working on.

You write the buyer questions yourself and group them into topics — Attensira does not sell you keyword volume, because a model answer has no search volume. What it gives you back is how each of those questions is answered today: who was named, which domains were linked, and how often, with the sample size attached.

It follows a brand profile and reusable Skills you define, which govern voice and rules. That is available on Growth and Business, and it changes how the agent writes — never what it measures.

As a pull request against the repository the page lives in, through a GitHub App you install into your own org. Each PR sits on a fresh branch, carries the `attensira` label, and names the page, the finding, and the reasoning in its body. The agent never merges and never pushes to your default branch, so an editor is always the last step.

Those tools score a draft against Google SERPs. Attensira measures whether models named you and linked you in answers it collected and stored, and then writes the edit and opens the PR. There is no content grade to chase — there is a mention rate, a citation rate, and a diff.