Plain-English questions
Ask “where are we weakest?” instead of building a filter.
Platform
Ask your data anything
The agent reads your project’s scans, citations and competitors. Ask it a question or run a skill, and get an answer you can check against the source data.
How it works
How agent works
01
Open the agent
Start a new chat in your project. It has access to that project’s data only.
02
Ask or pick a skill
Type a question, or run a prepared skill such as the weekly recap.
03
Get a grounded answer
Answers are built from your scans, citations and competitors, and reference the prompts they come from.
04
Turn it into work
Hand a finding to the content agent, or ask for a plan of next steps.
Capabilities
Everything in one place
Playbooks
Prepared workflows for visibility, citations, content and research.
weekly recap
A weekly summary of what changed, what moved and what to do next.
Gap finder
Lists prompts where competitors are named and you are not.
Source-aware
Answers can point to the citations and prompts behind a claim.
Chat history
Past chats are kept per project, grouped by date.
In depth
What you can do with it
Chat
Questions that would take five filters
Instead of clicking through tabs, ask for the comparison you want. The agent queries your project data and answers in a few lines.
- Works over your scans, citations and competitors
- Follow-up questions keep context
- History saved per project
Skills
Skills for the jobs you repeat
A skill is a prepared task with a fixed shape of output. Run the same briefing every Monday and compare it with last week’s.
- Categories: Popular, Visibility, Citations, Content, Research
- Consistent output you can compare week to week
- Start one from the empty chat screen
Handoff
From insight to a draft
When the agent finds a gap, hand it to a content agent in the same flow, so the insight does not get lost in a document.
- Create a brief from a finding
- Schedule the article on the calendar
- Track the prompt afterwards
Metrics
What we measure, defined Every number has one definition, so you can check it against the data.
- Grounded answer
- An answer built from your project’s scan, citation and competitor data rather than general knowledge.
- Skill
- A prepared agent task with a fixed output format, such as the weekly recap.
- Chat history
- Saved conversations for a project, so you can reopen an earlier analysis.
Related
Works with the rest of the platform
Prompt tracking
Track the prompts your buyers ask AI. Citeflare scans ChatGPT, Claude, Gemini, Perplexity and Google AI Overviews every day and records mentions, position and sentiment.
Competitors
Measure how often AI engines name you versus competitors. Share of voice, a heatmap by engine and competitor suggestions pulled from real answers.
Content agents
Turn a gap into a brief, an article and a published page. Citeflare’s content agents write, schedule and publish to WordPress or a webhook, and refresh existing pages with a diff.
Faq
Questions, answered
What data can the agent see?
Only the project you opened it in: its prompts, scan results, citations, competitors and content.
Can the agent be wrong?
Yes, like any language model. It works from your data and points at it, so check important claims against the source tables.
Does the agent change my data?
It reads data and can draft briefs. Publishing still needs your approval.
What is a skill?
A saved workflow, such as a weekly briefing or a content-gap search, that returns the same shape of answer each time.
Is the agent on every plan?
See the pricing page for what each plan includes.
See what AI says about you. Your first scan runs the day you sign up.
Monthly or annual plans. Cancel any time.