# Get Me
Source: https://docs.tryhikoo.com/api-reference/account/get-me
/api-reference/openapi.json get /api/v1/public/me
Return the user the API key is bound to.
# Audit Selected Pages
Source: https://docs.tryhikoo.com/api-reference/audits/audit-selected-pages
/api-reference/openapi.json post /api/v1/public/audits/pages
Audit a specific list of URLs on the website. All URLs must belong to the
website's domain. The number of URLs is capped by the plan's
``audit_pages_per_month_per_website`` quota.
# Audit Single Url
Source: https://docs.tryhikoo.com/api-reference/audits/audit-single-url
/api-reference/openapi.json post /api/v1/public/audits/url
Audit a single URL on the website. URL must belong to the website's domain.
# Create Audit
Source: https://docs.tryhikoo.com/api-reference/audits/create-audit
/api-reference/openapi.json post /api/v1/public/audits
Queue a new full-site audit for the target website.
# Get Audit
Source: https://docs.tryhikoo.com/api-reference/audits/get-audit
/api-reference/openapi.json get /api/v1/public/audits/{audit_id}
Retrieve a specific audit with its audited pages.
# Get Audit Page
Source: https://docs.tryhikoo.com/api-reference/audits/get-audit-page
/api-reference/openapi.json get /api/v1/public/audits/{audit_id}/pages/{page_id}
Detail of a single audit page, including every criterion (validated or not)
and the recommendations attached to each.
# Get Audits Quota
Source: https://docs.tryhikoo.com/api-reference/audits/get-audits-quota
/api-reference/openapi.json get /api/v1/public/audits/quota
Audit quota usage (used / limit / remaining) for the current billing
period. Call before ``hk_start_audit`` to avoid hitting the cap.
# List Audits
Source: https://docs.tryhikoo.com/api-reference/audits/list-audits
/api-reference/openapi.json get /api/v1/public/audits
List all audits for the target website (most recent first).
# Sync Audit
Source: https://docs.tryhikoo.com/api-reference/audits/sync-audit
/api-reference/openapi.json post /api/v1/public/audits/{audit_id}/sync
Queue a new audit that reuses the page set and criteria of a previously
completed audit. Cheap way to re-check the same surface after content changes.
# Get All Source Urls
Source: https://docs.tryhikoo.com/api-reference/citations/get-all-source-urls
/api-reference/openapi.json get /api/v1/public/citations/all-sources
Full list of source URLs cited in AI responses, paginated and sorted by
frequency. Unlike /top-sources this is not capped at 50 entries.
# Get Citation
Source: https://docs.tryhikoo.com/api-reference/citations/get-citation
/api-reference/openapi.json get /api/v1/public/citations/{citation_id}
Single citation by id: full content, sources, brand_mentions, prompt,
competitor ranks. ``brand_mentions`` is stripped when the website's plan
disables sentiment analysis.
# Get Citations Breakdown
Source: https://docs.tryhikoo.com/api-reference/citations/get-citations-breakdown
/api-reference/openapi.json get /api/v1/public/citations/breakdown
Per-match_type counts (brand_citation / site_mention / background_source)
over the requested period, with delta vs previous period and a daily
sparkline. Powers trend cards.
# Get Platforms Share Of Voice
Source: https://docs.tryhikoo.com/api-reference/citations/get-platforms-share-of-voice
/api-reference/openapi.json get /api/v1/public/citations/platforms
Visibility trend data per platform for the target website.
# Get Sentiment Summary
Source: https://docs.tryhikoo.com/api-reference/citations/get-sentiment-summary
/api-reference/openapi.json get /api/v1/public/citations/sentiment-summary
Aggregated sentiment per brand (tracked website + competitors +
unresolved mentions) over the requested period.
Plan-gated: 403 when ``sentiment_analysis_enabled=False`` on the
website's active subscription.
# Get Sentiment Trend
Source: https://docs.tryhikoo.com/api-reference/citations/get-sentiment-trend
/api-reference/openapi.json get /api/v1/public/citations/sentiment-trend
Sentiment counts bucketed over time (day or week).
Plan-gated. 403 when sentiment_analysis_enabled is false.
# Get Top Platforms
Source: https://docs.tryhikoo.com/api-reference/citations/get-top-platforms
/api-reference/openapi.json get /api/v1/public/citations/top-platforms
Top AI platforms by citation count for the target website.
# Get Top Sources
Source: https://docs.tryhikoo.com/api-reference/citations/get-top-sources
/api-reference/openapi.json get /api/v1/public/citations/top-sources
Top source domains (or URLs) appearing in AI citations for the target website.
# List Citations
Source: https://docs.tryhikoo.com/api-reference/citations/list-citations
/api-reference/openapi.json get /api/v1/public/citations
Paginated list of AI citations for the target website (full content, sources, competitors).
# Add Competitor
Source: https://docs.tryhikoo.com/api-reference/competitors/add-competitor
/api-reference/openapi.json post /api/v1/public/competitors
Add a competitor to the website. Plan quota (``competitors_limit``) is
enforced inside the manager.
# Delete Competitor
Source: https://docs.tryhikoo.com/api-reference/competitors/delete-competitor
/api-reference/openapi.json delete /api/v1/public/competitors/{competitor_id}
Remove a tracked competitor from the website.
# Dismiss Competitor Suggestion
Source: https://docs.tryhikoo.com/api-reference/competitors/dismiss-competitor-suggestion
/api-reference/openapi.json post /api/v1/public/competitors/suggestions/{suggestion_id}/dismiss
Dismiss a suggested competitor (hides it from future lists).
# Get Competitive Share Over Time
Source: https://docs.tryhikoo.com/api-reference/competitors/get-competitive-share-over-time
/api-reference/openapi.json get /api/v1/public/competitors/website-shares
Citation share of voice over time for the target website (vs all competitors).
# Get Competitor Citations
Source: https://docs.tryhikoo.com/api-reference/competitors/get-competitor-citations
/api-reference/openapi.json get /api/v1/public/competitors/citations
Citation counts per competitor for the requested period.
# Get Competitors Chart
Source: https://docs.tryhikoo.com/api-reference/competitors/get-competitors-chart
/api-reference/openapi.json get /api/v1/public/competitors/chart-stats
Share of voice chart data for the website and its competitors.
# Get Competitors Quota
Source: https://docs.tryhikoo.com/api-reference/competitors/get-competitors-quota
/api-reference/openapi.json get /api/v1/public/competitors/limit
Plan-based competitor quota for the website.
# Get Competitors Ranking
Source: https://docs.tryhikoo.com/api-reference/competitors/get-competitors-ranking
/api-reference/openapi.json get /api/v1/public/competitors
Competitor ranking (share of voice) for the target website.
# List Competitor Suggestions
Source: https://docs.tryhikoo.com/api-reference/competitors/list-competitor-suggestions
/api-reference/openapi.json get /api/v1/public/competitors/suggestions
List competitor candidates surfaced by the citation analyzer.
Active (non-dismissed, non-promoted) suggestions sorted by
``citation_count`` DESC. Plan-gated: 403 if competitor suggestions are
disabled on the website's plan.
# Promote Competitor Suggestion
Source: https://docs.tryhikoo.com/api-reference/competitors/promote-competitor-suggestion
/api-reference/openapi.json post /api/v1/public/competitors/suggestions/{suggestion_id}/promote
Promote a suggested competitor to a tracked Competitor row.
The plan's competitor quota is enforced inside the manager - a 403 is
raised if the website already has ``competitors_limit`` active rows.
# Update Competitor
Source: https://docs.tryhikoo.com/api-reference/competitors/update-competitor
/api-reference/openapi.json put /api/v1/public/competitors/{competitor_id}
Update a competitor's name, URL, or aliases.
# Get Content Gap
Source: https://docs.tryhikoo.com/api-reference/content-gaps/get-content-gap
/api-reference/openapi.json get /api/v1/public/content-gaps/{suggestion_id}
Full detail of a content-gap suggestion: brief, outline, reasoning,
similar competitor topic, closest existing page on the website.
# Get Content Gap Quota
Source: https://docs.tryhikoo.com/api-reference/content-gaps/get-content-gap-quota
/api-reference/openapi.json get /api/v1/public/content-gaps/quota
Monthly content-gap analysis quota for the website.
# Get Content Gap Summary
Source: https://docs.tryhikoo.com/api-reference/content-gaps/get-content-gap-summary
/api-reference/openapi.json get /api/v1/public/content-gaps/summary
Summary stats for content gap suggestions: counts by status, priority, etc.
# List Content Gaps
Source: https://docs.tryhikoo.com/api-reference/content-gaps/list-content-gaps
/api-reference/openapi.json get /api/v1/public/content-gaps
Paginated content-gap suggestions sorted by similarity score (lowest first
means biggest gaps).
# Dismiss Suggestion
Source: https://docs.tryhikoo.com/api-reference/health/dismiss-suggestion
/api-reference/openapi.json post /api/v1/public/health/suggestions/{suggestion_id}/dismiss
Dismiss a health suggestion so it stops appearing in the overview.
# Get Health
Source: https://docs.tryhikoo.com/api-reference/health/get-health
/api-reference/openapi.json get /api/v1/public/health
Global health (AI crawlers allowed, llms.txt, sitemap) + improvement suggestions.
# Get Health Refresh Quota
Source: https://docs.tryhikoo.com/api-reference/health/get-health-refresh-quota
/api-reference/openapi.json get /api/v1/public/health/refresh-quota
Plan-based quota usage for health refreshes in the current billing period.
# Get Suggestion Pages
Source: https://docs.tryhikoo.com/api-reference/health/get-suggestion-pages
/api-reference/openapi.json get /api/v1/public/health/suggestions/{suggestion_id}/pages
Paginated list of indexed pages affected by a health suggestion.
# Get Suggestion Regen Quota
Source: https://docs.tryhikoo.com/api-reference/health/get-suggestion-regen-quota
/api-reference/openapi.json get /api/v1/public/health/suggestions/regenerate-quota
Plan-based quota usage for suggestion regenerations in the current billing period.
# Refresh Health
Source: https://docs.tryhikoo.com/api-reference/health/refresh-health
/api-reference/openapi.json post /api/v1/public/health/refresh
Force a re-check of robots.txt, llms.txt, and sitemap. Quota-limited
by the plan's ``health_refresh_monthly_website_limit``.
# Regenerate Suggestions
Source: https://docs.tryhikoo.com/api-reference/health/regenerate-suggestions
/api-reference/openapi.json post /api/v1/public/health/suggestions/regenerate
Regenerate health suggestions from currently indexed pages.
Plan-limited by ``suggestion_regen_monthly_website_limit``.
# Restore Suggestion
Source: https://docs.tryhikoo.com/api-reference/health/restore-suggestion
/api-reference/openapi.json post /api/v1/public/health/suggestions/{suggestion_id}/restore
Restore a previously dismissed health suggestion.
# Get Indexation Status
Source: https://docs.tryhikoo.com/api-reference/indexation/get-indexation-status
/api-reference/openapi.json get /api/v1/public/indexation/status
Latest indexation job status (running, completed, failed) plus progress counters.
# Get Indexation Summary
Source: https://docs.tryhikoo.com/api-reference/indexation/get-indexation-summary
/api-reference/openapi.json get /api/v1/public/indexation/summary
Aggregated stats on indexed pages: total, by category, health pass rate,
pages with errors or warnings, last crawl timestamp.
# Get Page
Source: https://docs.tryhikoo.com/api-reference/indexation/get-page
/api-reference/openapi.json get /api/v1/public/indexation/pages/{page_id}
Detail of one indexed page: URL, title, language, category, last crawled at, errors.
# Index Url
Source: https://docs.tryhikoo.com/api-reference/indexation/index-url
/api-reference/openapi.json post /api/v1/public/indexation/single-page
Trigger a re-crawl of a single URL on the website. The URL must belong
to the website's domain. Returns a job handle.
# List Page Errors
Source: https://docs.tryhikoo.com/api-reference/indexation/list-page-errors
/api-reference/openapi.json get /api/v1/public/indexation/errors
Most common health-check errors across the website's indexed pages.
# List Pages
Source: https://docs.tryhikoo.com/api-reference/indexation/list-pages
/api-reference/openapi.json get /api/v1/public/indexation/pages
Paginated list of indexed pages, with filters on category, health,
errors, warnings, language, and URL/title search.
# REST API overview
Source: https://docs.tryhikoo.com/api-reference/introduction
Programmatic access to your Hikoo data: citations, prompts, competitors, audits, health and more.
The Hikoo REST API exposes the same data you see in the dashboard, and the same surface the [MCP server](/guides/mcp/overview) relays. Use it to pull GEO analytics into your own tools, scripts, or pipelines.
```
https://api.tryhikoo.com/api/v1/public
```
## Authentication
Every request is authenticated with a personal API key passed as a Bearer token:
```bash theme={null}
curl "https://api.tryhikoo.com/api/v1/public/me" \
-H "Authorization: Bearer hk_your_api_key"
```
To create a key, go to **Settings > API keys** in the [dashboard](https://app.tryhikoo.com), click **Create key**, and copy it immediately: it starts with `hk_` and is only shown once.
The same key works for both the REST API and the [MCP server](/guides/mcp/setup). It is tied to your user account and covers every workspace and website you belong to.
Treat keys like passwords: never commit them to a repository or expose them in client-side code. Revoke and recreate a key if it leaks.
## Targeting a website
Most endpoints operate on a single website and take a `website_id` query parameter. To discover your websites:
1. Call `GET /workspaces/with-websites` to list every workspace and the websites inside.
2. Pick the `id` of the website you want.
3. Pass it as `?website_id=` to the data endpoints.
```bash theme={null}
curl "https://api.tryhikoo.com/api/v1/public/citations?website_id=123&page=1" \
-H "Authorization: Bearer hk_your_api_key"
```
Data endpoints require an **active subscription** on the target website; without one they return `403` with an explanatory message.
## Pagination
List endpoints accept `page` (1-based) and `page_size` parameters and return a consistent envelope:
```json theme={null}
{
"items": [],
"total": 240,
"page": 1,
"page_size": 20,
"total_pages": 12
}
```
## Rate limits
Requests are limited **per API key** to:
| Window | Limit |
| ---------- | -------------- |
| Per minute | 60 requests |
| Per hour | 5,000 requests |
Exceeding a limit returns `429 Too Many Requests`. Note that MCP traffic using the same key counts against the same budget.
## Errors
The API uses standard HTTP status codes. Error responses carry a `detail` field:
```json theme={null}
{ "detail": "Invalid or expired API key" }
```
| Status | Meaning |
| ------ | -------------------------------------------------------------- |
| `401` | Missing, invalid, or expired API key |
| `403` | No active subscription on the website, or a plan-gated feature |
| `404` | Resource not found, or not visible to your account |
| `422` | Invalid parameters (the response details which ones) |
| `429` | Rate limit exceeded |
## REST API or MCP?
Both expose the same data with the same key.
* **REST API**: your own code, scheduled jobs, BI pipelines, integrations.
* **MCP server**: AI assistants like Claude exploring your data conversationally. See the [MCP setup guide](/guides/mcp/setup).
# Create Prompt
Source: https://docs.tryhikoo.com/api-reference/prompts/create-prompt
/api-reference/openapi.json post /api/v1/public/prompts
Create a new prompt. Fails if the website's prompt quota is reached.
# Delete Prompt
Source: https://docs.tryhikoo.com/api-reference/prompts/delete-prompt
/api-reference/openapi.json delete /api/v1/public/prompts/{prompt_id}
Delete a prompt.
# Get Prompt Citations
Source: https://docs.tryhikoo.com/api-reference/prompts/get-prompt-citations
/api-reference/openapi.json get /api/v1/public/prompts/{prompt_id}/citations
Paginated citations tied to a specific prompt.
# Get Prompt Detail
Source: https://docs.tryhikoo.com/api-reference/prompts/get-prompt-detail
/api-reference/openapi.json get /api/v1/public/prompts/{prompt_id}/detail
Aggregate payload for prompt drill-down: visibility + 7d delta, best
rank with platform, coverage X/N, 30-day sparkline, per-platform coverage
grid, and the latest N answers in a single round trip.
# Get Prompt Stats
Source: https://docs.tryhikoo.com/api-reference/prompts/get-prompt-stats
/api-reference/openapi.json get /api/v1/public/prompts/{prompt_id}/stats
Aggregated visibility stats for a prompt across every citation.
# Get Prompts Quota
Source: https://docs.tryhikoo.com/api-reference/prompts/get-prompts-quota
/api-reference/openapi.json get /api/v1/public/prompts/quota
Return the current prompt count, subscription limit, and remaining slots.
Use this BEFORE bulk-creating prompts to avoid hitting the quota mid-batch.
# List Prompts
Source: https://docs.tryhikoo.com/api-reference/prompts/list-prompts
/api-reference/openapi.json get /api/v1/public/prompts
List prompts grouped by topic, each with its visibility score.
# Update Prompt
Source: https://docs.tryhikoo.com/api-reference/prompts/update-prompt
/api-reference/openapi.json put /api/v1/public/prompts/{prompt_id}
Update an existing prompt.
# Get Citation Search Queries
Source: https://docs.tryhikoo.com/api-reference/search-queries/get-citation-search-queries
/api-reference/openapi.json get /api/v1/public/search-queries/by-citation/{citation_id}
All search query variants attached to a single citation. Useful for debugging
which AI sub-question generated a citation.
# Get Prompt Search Queries
Source: https://docs.tryhikoo.com/api-reference/search-queries/get-prompt-search-queries
/api-reference/openapi.json get /api/v1/public/search-queries/by-prompt/{prompt_id}
Unique search queries variants executed for a specific prompt.
# List Search Queries
Source: https://docs.tryhikoo.com/api-reference/search-queries/list-search-queries
/api-reference/openapi.json get /api/v1/public/search-queries
Unique AI search query variants executed for the website, aggregated and paginated.
# Get Stats
Source: https://docs.tryhikoo.com/api-reference/source-intelligence/get-stats
/api-reference/openapi.json get /api/v1/public/source-intelligence/stats
Aggregated stats: counts by source type and by platform.
# List Sources
Source: https://docs.tryhikoo.com/api-reference/source-intelligence/list-sources
/api-reference/openapi.json get /api/v1/public/source-intelligence/sources
List suggested sources for the target website, ordered by citation count desc.
# Create Topic
Source: https://docs.tryhikoo.com/api-reference/topics/create-topic
/api-reference/openapi.json post /api/v1/public/topics
Create a new topic for the target website.
# Delete Topic
Source: https://docs.tryhikoo.com/api-reference/topics/delete-topic
/api-reference/openapi.json delete /api/v1/public/topics/{topic_id}
Delete a topic from the website.
# List Topics
Source: https://docs.tryhikoo.com/api-reference/topics/list-topics
/api-reference/openapi.json get /api/v1/public/topics
List all topics tracked for the target website.
# Update Topic
Source: https://docs.tryhikoo.com/api-reference/topics/update-topic
/api-reference/openapi.json put /api/v1/public/topics/{topic_id}
Rename a topic.
# Get Subscription Details
Source: https://docs.tryhikoo.com/api-reference/website/get-subscription-details
/api-reference/openapi.json get /api/v1/public/website/subscription
Plan info for the website: plan tier, product name, status, trial/expiry
dates, AI platforms covered, feature flags, and quotas. Agents use this to
tell the user which features require an upgrade.
Restricted to workspace OWNER and ADMIN. Members get 403 because billing
info (plan, expiry, quotas) is sensitive.
# Get Website
Source: https://docs.tryhikoo.com/api-reference/website/get-website
/api-reference/openapi.json get /api/v1/public/website
Return the website detail (subscription, plan, settings) for the given website_id.
# List Websites
Source: https://docs.tryhikoo.com/api-reference/websites/list-websites
/api-reference/openapi.json get /api/v1/public/websites
List websites the user can access, paginated.
Each item includes ``has_active_subscription`` so the MCP / Claude can
tell the user which sites are queryable. Sites without an active sub
are still listed (so Claude can mention them) but their data endpoints
return 403.
# List Workspaces
Source: https://docs.tryhikoo.com/api-reference/workspaces/list-workspaces
/api-reference/openapi.json get /api/v1/public/workspaces
List every workspace the user is a member of (excludes soft-deleted).
# List Workspaces With Websites
Source: https://docs.tryhikoo.com/api-reference/workspaces/list-workspaces-with-websites
/api-reference/openapi.json get /api/v1/public/workspaces/with-websites
Workspaces the user belongs to, each with its websites nested inside.
One-shot discovery: the caller gets the full tree (workspaces → websites
+ active-subscription flag) in a single call so an agent can pick the
right website without follow-up requests.
# How Hikoo works
Source: https://docs.tryhikoo.com/concepts/how-hikoo-works
The core concepts behind Hikoo: prompts, citations, share of voice, and AI readiness.
Hikoo answers one question: when people ask AI assistants about your space, does your brand show up, and what can you do about it. This page explains the ideas that everything else builds on.
## From prompts to citations
### You define prompts
A **prompt** is a natural-language question a real user might ask an AI assistant, for example "best project management tool for startups". You create prompts, import them, or let Hikoo generate them. See [Prompts](/guides/prompts).
### Hikoo asks the AI assistants
Every day, Hikoo sends your prompts to the AI assistants and records their full answers. Behind each answer the assistant runs its own **search queries**, which Hikoo also captures. See [Search queries](/guides/search-queries).
### Hikoo records citations
A **citation** is one instance of an assistant mentioning or linking your brand (or a competitor) in an answer. Hikoo matches, ranks, and analyzes each answer to know who was cited, in what position, and how positively. See [Citations](/guides/citations).
## The metrics you will see
| Term | What it means |
| -------------------- | --------------------------------------------------------------------------- |
| **Visibility score** | How often your site is cited relative to competitors. |
| **Share of voice** | Your share of all citations across the tracked brands. |
| **Rank** | Your position in a given answer (for example rank 2 of 6). |
| **Sentiment** | Whether an answer portrays your brand positively, neutrally, or negatively. |
| **Citation rate** | How often a prompt produces a citation of your site. |
## Platforms Hikoo tracks
Hikoo tracks visibility across the major AI assistants and answer engines:
Which platforms you can track is set by your plan.
## The two sides of GEO
Winning AI citations takes two things, and Hikoo covers both.
AI readiness: your site must let AI crawlers in and expose clean, structured content. This is the **Analyzer** (audit, health, pages).
Coverage: you need content and third-party presence for the topics AI answers. This is **Elevate** (content gaps) and **Battlemap** (competitors, source intelligence).
## How the app is organized
The dashboard groups features into modules. This documentation follows the same structure.
| Module | What it covers |
| ------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Overview** | The [dashboard](/guides/dashboard) home and your [marketing plan](/guides/marketing-plan). |
| **Spotlight** | [Citations](/guides/citations), [prompts](/guides/prompts), [topics](/guides/topics), [search queries](/guides/search-queries), [sentiment](/guides/sentiment), and [agent analytics](/guides/agent-analytics). |
| **Battlemap** | [Competitors](/guides/competitors) and [source intelligence](/guides/source-intelligence). |
| **Elevate** | [Content gaps](/guides/elevate) where competitors get cited and you do not. |
| **Analyzer** | [Audit](/guides/audit), [health](/guides/health), and [pages](/guides/pages). |
## Everything is scoped to a website
Every metric belongs to a specific **website**. Use the site switcher in the sidebar to change the active site, and the shared filter bar (time range, platform, topic, market) to slice any view. Websites live inside [workspaces](/guides/workspaces), which hold your team and your subscription.
Open the dashboard guide to see these concepts in the product.
# Agent analytics
Source: https://docs.tryhikoo.com/guides/agent-analytics
See real AI agents and crawlers fetching pages on your own site
**Agent analytics** shows the real AI agent and crawler traffic hitting your own site. It tells you which AI vendors crawl you, which URLs they read, and how that activity changes over time.
## How this differs from citations
Citations and agent analytics answer two different questions:
* **Citations** measure whether AI answers cite you. See [Citations](/guides/citations).
* **Agent analytics** measures whether AI bots are actually fetching your pages.
A page can be crawled without being cited, and cited without being freshly crawled, so it helps to watch both.
## What the page shows
Cards across the top give you the headline numbers:
* **Total visits**
* **Unique agents**
* **Unique pages**
* **Top vendor**
* **Top agents**
You can filter by **vendor**, read a **visits over time** chart, and drill into a per-URL and per-folder table to see exactly which pages the agents read.
## Set up agent analytics
### Open the configuration panel
Open the configuration and connection panel for agent analytics.
### Generate a tracking token
Generate a site tracking token for your website.
### Install the tracking script
Add the provided tracking script to your site so incoming agent traffic is recorded.
### Wait for ingestion
Once traffic is ingested, the dashboard populates with your agent visits.
## Availability
Agent analytics is available on some plans. Once it is on your plan, complete the setup above to start collecting data.
## Related
Check whether your site is readable by AI crawlers.
See the pages agents are reading, page by page.
# API keys
Source: https://docs.tryhikoo.com/guides/api-keys
Create and manage keys that authenticate the Hikoo REST API and MCP server.
API keys authenticate the Hikoo public REST API and the Hikoo MCP server. A key proves who you are so those interfaces can read the data you have access to.
## How a key is scoped
A key is scoped to your user account. One key can reach every website in the workspaces you belong to, subject to each site's subscription. This replaces the older per-website key model, so you no longer need a separate key for each site.
Access still follows subscriptions. A key can address a site only when that site has an active subscription.
## Create a key
Open Settings, then the API keys section.
Give it a descriptive name so you can tell your keys apart later.
Add an expiry date if you want the key to stop working automatically, or leave it open.
The full key is shown only once and starts with `hk_`. Copy it immediately and store it somewhere safe. If you lose it, create a new one.
## Manage keys
From the API keys section you can list your keys and see their names and status. Revoke or delete a key to disable it immediately. Any client using that key stops working at once, so rotate before you revoke if a client still needs access.
## Install cards
The API keys section also gives you two install cards:
Get the steps to connect an AI client to your data.
Get the steps to call the Hikoo REST API directly.
## Keep keys safe
Treat keys like passwords. Never commit them to source control or paste them where others can read them. If a key leaks, rotate it: create a new key, update your clients, then revoke the old one.
## Related pages
Wire a key into an AI client.
See the tools the MCP server exposes.
Where the API keys section lives.
# Audit
Source: https://docs.tryhikoo.com/guides/audit
Run an on-demand AI-readiness evaluation of your site or specific pages and track your score over time
The audit runs an on-demand AI-readiness evaluation of your website. It is part of the **Analyzer** module. Hikoo scores each page against a weighted catalog of criteria, then rolls the page scores up into an overall score you can track over time.
Use the audit when you want a page-by-page verdict on how well your content is prepared to be found and cited by AI assistants. For a site-wide technical readiness overview, see [Health](/guides/health).
## Running an audit
Click **New audit** to open the dialog. You have three modes.
On the **Full site** tab, Hikoo audits every page it has discovered for the selected website. Use this for a complete picture of your AI readiness.
Audit one specific URL. Use this to check a single page after you have changed it.
On the **Select pages** tab, choose a set of pages to evaluate. Use this when you want to focus on a section of your site.
The audit runs in the background. A progress banner shows its status while it is in process. You can leave the page and come back. When the run finishes, use **Sync** to pull in the results, or wait for the audit to update on its own.
Audits are scoped to the website selected in the site switcher. Switch sites in the sidebar to audit a different property.
## How pages are scored
Each page is scored against a weighted criteria catalog that mixes static checks and LLM checks. Static checks read your page directly. LLM checks use a model to judge qualities that are hard to measure mechanically, such as whether a heading reads like a real question.
Criteria are scoped by page category, so a blog post is judged by different rules than a product page. Categories include home, blog post, product, e-commerce product page, service, FAQ, and landing page. A set of site-wide `GLOBAL` checks applies to the property as a whole.
### Example criteria
* `llms.txt` exists and is well formed.
* `robots.txt` exists and allows AI crawlers.
* `sitemap.xml` exists.
* Article schema or another relevant schema is present as JSON-LD.
* The page has a clean H1 to H2 to H3 heading hierarchy.
* The page loads in under 3 seconds on mobile.
* No content is hidden behind JavaScript.
* Headings are descriptive and question-style.
* Metadata is strong.
* E-commerce product pages meet their page-specific requirements.
## Audit results
When an audit completes, the summary shows:
* **Overall score**, displayed as a ring.
* **Pages analyzed**, the total number of pages evaluated.
* **Average page score**, the mean score across those pages.
* **Assessment**, a plain-language rating of Excellent, Good, Needs work, or Poor.
* **Pages with errors**, so you can jump straight to what needs attention.
## Page-level detail
Open any page in the audit to see its full breakdown. Each page lists every criterion with a pass or fail result and a recommendation for the ones that fail. Criteria are grouped into areas so you can see where the weakness is, covering performance, accessibility, SEO, and best practices. Score charts show how the page performs across those areas.
## Audit actions
From an audit you can:
* **Sync** to pull in the latest results.
* **Export** the audit.
* **Archive** an audit you want to keep but set aside.
* **Delete** an audit you no longer need.
## Quotas
Your plan determines your audit quotas. Three limits apply:
* **Audits per month**, how many audit runs you can start in a billing period.
* **Audited pages per month**, the total pages you can evaluate across all runs.
* **Page limit per audit**, the largest number of pages a single run can cover.
A remaining counter on the Audit page shows what you have left in the current period.
## Related guides
Check your site-wide technical readiness for AI discovery.
Browse your indexed pages and their per-page audit scores.
Close content gaps and generate articles from your findings.
Turn audit findings into a prioritized task board.
# Citations
Source: https://docs.tryhikoo.com/guides/citations
Track every time an AI assistant mentions or links your site in an answer
A **citation** is one instance of an AI assistant mentioning or linking your site in an answer to a tracked prompt. The Citations page collects those instances, benchmarks them against competitors, and lets you filter down to a single answer to see exactly how you were referenced.
## Citations by platform
At the top of the page, the **Citations by platform** chart breaks your citation volume down by AI platform over time. Alongside it, insight callouts summarize the movement so you do not have to read the chart line by line. They highlight your most cited platform, the fastest growing one, any platform that is declining, and any newcomer platform that just started citing you.
## All citations
Below the chart, the **All citations** table lists every recorded citation. Each row gives you:
| Column | What it shows |
| ----------------- | ---------------------------------------------- |
| Date | When the citation was collected |
| AI platform | The assistant that produced the answer |
| Cited | Whether your site was cited in that answer |
| Sentiment | How the answer portrayed your brand |
| Competitors cited | Which rivals appeared in the same answer |
| Sources | The domains and pages the answer pulled from |
| Rank | Where your site landed among the cited sources |
Select any row to open its detail view.
## Filters
A filter bar sits above the table so you can narrow the list to the citations you care about. You can filter by:
* **Platform**, **time range**, **topic**, and **market**
* **Competitor**, to see answers where a specific rival appeared
* **Sentiment**, across positive, neutral, and negative
* **Rank**, using top-N, equal to N, at least N, or between two values
* **Cited only**, to hide answers where your site was not referenced
* **Match types**, across brand citation, site mention, and background source
Active filters appear as chips, and a **Clear all** action resets them at once.
## Citation detail
The citation detail view shows the full AI answer so you can read it in context. It marks whether your site was cited, lists the competitors cited in the same answer, and separates the sources into **Mentioned** and **Background**. A rank badge reads **Rank X of total** so you can see your standing among all cited sources at a glance.
## Citation quota
Your plan buys a number of AI-response "runs" that Hikoo spends collecting fresh answers. The citation quota panel shows runs used and remaining and your renewal date. It also gives you control over how those runs are spent:
* An **auto-distribution** toggle lets Hikoo spread runs across your platforms for you.
* A manual distribution editor lets you set the split across platforms yourself.
* **Launch now** triggers a collection immediately instead of waiting for the daily run.
Which platforms you can track, and how many runs you get, are set by your plan.
## Sources
The **Sources** sub-page lists the third-party domains and pages that AI answers pulled from when they cited your space. The **All sources** table groups these by source class so you can see which kinds of sites feed the answers you appear in. Use it to find the pages worth earning a mention on.
## How collection works
Hikoo automatically collects fresh citations every day by querying the real AI assistants with your prompts. It then matches each answer against your site and competitors, ranks where you landed, and analyzes the sentiment and sources. Each answer spends one run from your plan. You can also spend a run on demand with **Launch now** on the citation quota panel.
Export is available on plans that include it.
## Related
The tracked questions that produce your citations.
Benchmark your citations against rival brands.
See how AI answers portray your brand.
The sub-questions AI assistants search on your behalf.
# Competitors
Source: https://docs.tryhikoo.com/guides/competitors
Benchmark your AI citation performance against competitors with the Battlemap
The Competitors page is your Battlemap. It shows how your citation performance compares to the other brands you track, so you can see where you lead, where you are behind, and exactly where to act next.
Everything on this page respects the shared filter bar. Change the **time range**, **platform**, or **market** and every number, chart, and table updates to match.
## Your position at a glance
The top of the page answers one question first. Are you winning or losing?
* A **share of voice** donut splits every citation across you and your tracked competitors, so you can read your slice of the total instantly.
* KPI tiles summarize your citations, average position, and share.
* A headline states where you stand in plain language, for example that you are **leading** or that you are **behind by N points**.
Share of voice is your share of all citations recorded across you and your competitors. Read more in [Citations](/guides/citations).
## Rankings table
A ranked table lists every domain you track, yours included, so you can compare like for like.
| Column | Description |
| -------------------- | ---------------------------------------------------- |
| **Rank** | Position across all tracked domains. |
| **Competitor** | The brand and its domain. |
| **Citations** | Total AI citations recorded in the selected range. |
| **Average position** | The typical placement of the brand when it is cited. |
| **Share** | Percentage of all citations captured by the brand. |
Your own site is highlighted in the table so you can find it without scanning.
## Managing competitors
You can add competitors during onboarding or at any time from this page.
### Add a competitor
Click **Add competitor**, then enter the brand name and its URL. Hikoo starts tracking their citations alongside yours.
### Edit and add aliases
Open a competitor to edit it. You can add **aliases**, the alternative names a brand is known by. Add as many as you need so citations that use a different spelling or a sub-brand still count toward the right competitor.
### Remove a competitor
Remove any competitor you no longer want to track. This frees a slot against your plan limit.
The number of competitors you can track is set by your plan. The page shows your usage as **X of limit** so you always know how many slots remain.
## Competitor suggestions
Hikoo watches your citations and surfaces brands that keep appearing next to you but that you are not yet tracking. These appear on the **Competitor suggestions** sub-page.
* Each suggestion shows its citations and when it was **last seen**.
* **Promote** a suggestion to start tracking it as a full competitor.
* **Dismiss** a suggestion to remove it from the list.
A badge in the sidebar shows how many suggestions are waiting for you.
Competitor suggestions are available on some plans.
## Competitor detail
Open any competitor to see a full profile of how they earn AI citations.
* A **trend** of their citations over time.
* Their **share of voice** and **average position**.
* Their total **citations**.
* **Top topic ranks**, so you can see the themes where they are strongest.
* A narrative **summary** that explains their position in words.
* The **sources** that cite them.
Use this view to understand not just that a competitor is ahead, but why.
## Movement and momentum
Rankings tell you where things stand today. The movement views tell you which way they are heading.
* **Velocity** shows how fast each brand is gaining or losing citation share.
* **Top movers** groups brands into **risers**, **fallers**, and **emergent** newcomers.
* **Quick wins** lists recoverable topics where a small effort can regain ground, each paired with a playbook.
* **Lost ground** collects the topics where you have slipped behind.
## Lost ground detail
Open a Lost ground topic to enter the interactive Battlemap. It walks you through a guided flow: you are losing here, here is how to fix it, and did it work.
Inside a Lost ground topic you see:
* **Brand ranks** for the topic, so you can see who is ahead of you.
* **Fix plays**, the concrete actions to close the gap.
* **Fix-page rows** that point to the specific pages to improve.
* **Outreach angles** that suggest where to earn new mentions.
Work through the plays, then return to check whether your rank recovered.
## Citation share over time
A historical chart tracks how each brand's citation share has moved across your selected range. Use it to confirm whether a change you made is holding, or whether a competitor is closing in.
## Where to go next
See the individual AI answers behind every number on this page.
Find the third-party sources that feed competitor citations.
Turn competitor gaps into content opportunities you can act on.
# Dashboard
Source: https://docs.tryhikoo.com/guides/dashboard
Get a bird's-eye view of your AI search visibility.
The dashboard is your home screen. It summarizes how the selected website performs across AI platforms, and it is the starting point for every other module.
## Overview cards
When you open the dashboard you see:
* **Visibility trend.** A chart of how often your site has been cited over time.
* **Competitor ranking.** Your site and its competitors ranked by citation count.
* **Citation share.** The share of citations going to you versus competitors.
* **Top source domains.** The domains and pages AI assistants pull from, with a toggle between domains and URLs.
* **Top platforms.** Which AI assistants cite you most often.
## First-wave banner
For a brand-new website, a first-wave banner tracks the initial analysis as it moves through generating prompts, querying the AI assistants, and indexing your pages. Data fills in as each stage completes.
## Shared filters
A filter bar at the top drives every chart. It is shared across the app, so what you set here carries into Citations, Competitors, and the other pages.
| Filter | Description |
| -------------- | --------------------------------------------------- |
| **Time range** | All time, 1 year, 1 month, 1 week, or 1 day. |
| **Platform** | A single AI platform or all platforms. |
| **Topic** | Focus on prompts in one [topic](/guides/topics). |
| **Market** | Focus on a geographic market, for example US or FR. |
Filters persist as you navigate, so you do not have to reapply them.
## Switching websites
If you track more than one website, use the **site switcher** in the sidebar to jump between them. Sites are grouped by workspace, and the whole dashboard updates to the site you pick. Sites without an active subscription prompt you to choose a plan first.
## Next steps
Dive into individual citation records.
See detailed competitive benchmarks.
# Elevate
Source: https://docs.tryhikoo.com/guides/elevate
Find the topics and questions competitors get cited for in AI answers but you do not
Elevate is the content-gap engine. It surfaces the topics and questions that competitors get cited for in AI answers, but that you do not. Each gap is a piece of content you could create to start earning those citations yourself.
## The gap list
Elevate presents a list of gap suggestions. Each one carries a **priority** and a **content type** so you can decide what to work on first.
* **Priority** is one of **high**, **medium**, or **low**.
* **Content type** describes the format to create: article, blog post, landing page, guide, how-to, FAQ, listicle, comparison, playbook, or checklist.
Open any suggestion to see the reasoning behind it:
* **Why to create it**, the case for filling this gap.
* A **content brief** to guide the writing.
* **Context** around where the gap comes from.
* The **original query** that exposed it.
* **Your closest existing page**, so you can decide whether to write something new or improve what you already have.
## Working a suggestion
Each suggestion has its own actions:
| Action | What it does |
| ------------------- | ----------------------------------------- |
| **Accept** | Commit to creating this content. |
| **Dismiss** | Set the suggestion aside. |
| **Mark as written** | Record that you published content for it. |
| **Delete** | Remove the suggestion entirely. |
When you click **Mark as written**, a dialog opens. Paste the URL you published, and optionally run an audit on that page at the same time so you can check how it scores.
Suggestions move through a clear status flow:
**New** to **Accepted** to **Written**, or **Dismissed** at any point.
## How Elevate finds gaps
Hikoo compares your own pages against the topics competitors get cited for, matching on meaning rather than exact wording. Where competitors are cited on a topic and you have no strong page, a gap appears.
### Prepare your pages
Open Elevate and let Hikoo prepare your pages for comparison.
### Run the analysis
Start the analysis. It runs in the background, so you can leave the page and come back.
### Review your gaps
When it finishes, your gap suggestions are ready to work through.
Results are best when your site is fully indexed and you have added relevant competitors. Index more pages from [Pages](/guides/pages) and add rivals in [Competitors](/guides/competitors) before you run the analysis.
Elevate runs against a monthly quota per website. Your plan determines how many analysis runs you get each month.
## Where to go next
Turn accepted gaps into scheduled tasks you can track.
Pair content gaps with the sources that would cite you.
Index more of your site so gap detection has more to compare.
Add the rivals whose coverage defines your gaps.
# Free tools
Source: https://docs.tryhikoo.com/guides/free-tools
Public tools on tryhikoo.com that check a site's AI readiness with no account.
Hikoo offers free, public tools on the marketing site at [tryhikoo.com](https://tryhikoo.com) that need no account. They are a fast way to sanity-check a site's AI readiness before you sign up. Each one runs on demand and returns a result in your browser.
## The tools
Enter a URL and get an AI-visibility audit. Entering your email unlocks the detailed report. This tool is rate-limited.
Generate and download an `llms.txt` file for your domain root.
Read a site's `robots.txt` and report, for each known AI crawler, whether it is allowed or blocked.
See what an AI crawler sees when it fetches a page without running JavaScript. It reports word, heading, and link counts, shows the raw extracted content, and warns if content is hidden behind JavaScript.
Convert any page into clean, LLM-ready markdown to copy or download.
The free tools give you a snapshot. Once you sign up, the in-app product tracks all of this continuously across every AI platform, so you see how visibility changes over time rather than at a single moment.
## Related pages
Continuous AI-readiness monitoring inside the app.
Full GEO and SEO audits for a tracked site.
Set up your first tracked website in Hikoo.
# Website health
Source: https://docs.tryhikoo.com/guides/health
See whether your site is technically ready to be discovered and cited by AI
Health is a site-wide GEO health overview. It answers one question: is your site technically ready to be discovered and cited by AI. It is part of the **Analyzer** module.
Health is distinct from the per-page [audit](/guides/audit). The audit scores individual pages. Health looks at the property as a whole, covering the files and crawler rules that decide whether AI assistants can reach your content at all.
## Global checks
Hikoo runs three site-wide checks and auto-detects the relevant URLs on your domain.
| Check | What it looks at |
| ------------------------ | ----------------------------------------------------------------------------------------------------------------------- |
| **robots.txt AI access** | Whether your `robots.txt` allows AI crawlers. This is evaluated per crawler, so you see exactly which ones are allowed. |
| **llms.txt** | Whether your site publishes an `llms.txt` file. |
| **sitemap.xml** | Whether a `sitemap.xml` is present so crawlers can discover your pages. |
## Resources editor
The resources editor lets you view and edit the three files that drive your AI readiness: your sitemap, your `llms.txt`, and your `robots.txt`. Hikoo shows the auto-detected version, and you can provide a custom one.
If you do not have an `llms.txt` yet, use the free `llms.txt` generator. See [Free tools](/guides/free-tools).
## Crawlers table
The crawlers table shows which AI crawlers are allowed or blocked on your site. Each crawler carries its importance, its category, and its purpose, so you can decide which ones matter to you.
* **Importance** is critical, high, or medium.
* **Category** is search, user, or training.
* **Purpose** describes what the crawler does.
Tracked crawlers include:
| Crawler | Owner |
| ------------------------------------------------ | ----------------- |
| GPTBot, OAI-SearchBot, ChatGPT-User | OpenAI |
| ClaudeBot, anthropic-ai, Claude-Web, Claude-User | Anthropic |
| Google-Extended, Gemini-Deep-Research | Google |
| PerplexityBot, Perplexity-User | Perplexity |
| Bingbot | Microsoft Copilot |
| MistralAI-User | Mistral |
| Bytespider | ByteDance |
| Amazonbot | Amazon |
| meta-externalagent | Meta |
| DuckAssistBot | DuckDuckGo |
Blocking a search or user crawler can stop an AI assistant from reaching your content when it answers a user. Review the importance column before you block a crawler.
## Suggestions
Health includes prioritized suggestions. These cover page improvements and new-content ideas, ordered so you can start with the ones that matter most. For each suggestion you can:
* **Regenerate** the full set after you change your site.
* **Dismiss** a suggestion you do not want, and **restore** it later.
* **View the affected pages** for a suggestion, so you can see where it applies.
## Indexation controls
Indexation is started and managed from the Health page. From here you can:
* **Start indexation** to have Hikoo discover and index your pages.
* **Re-index errored pages** to retry the pages that failed.
* **Add a single page** to the index.
A crawl-progress banner shows the status while indexation runs. The pages that come out of this process appear on the [Pages](/guides/pages) inventory.
## Refreshing health data
Health data is collected when you add a website. After that, Hikoo refreshes it automatically, and you can refresh it on demand at any time.
Refreshes and suggestion regenerations count against monthly quotas set by your plan. The remaining counts are shown on the page.
## Related guides
Browse the pages Hikoo has indexed for your site.
Score individual pages against the AI-readiness criteria.
Generate an `llms.txt` file for your domain.
# Marketing plan
Source: https://docs.tryhikoo.com/guides/marketing-plan
Turn your Hikoo findings into an actionable GEO task board
The marketing plan turns your Hikoo findings into an actionable GEO task board. It gathers the work your audits, sources, and content gaps imply, then lets you schedule and track it in one place.
The marketing plan is available on some plans.
## Views
The marketing plan has two views:
* A **calendar view**, which lays tasks out by date so you can plan your week.
* A **list view**, which shows tasks as a flat, orderable list.
## Tasks
Every task carries a set of attributes so you always know what it is, how urgent it is, and when it is due.
| Attribute | Values |
| ------------ | -------------------------------------------- |
| **Category** | Technical SEO, content, outreach, or custom. |
| **Status** | To do, in progress, or done. |
| **Priority** | 1 to 3. |
| **Effort** | The work a task takes. |
| **Duration** | How long a task should run. |
| **Due date** | When a task is due. |
## Generating tasks
Hikoo can propose tasks for you. It builds prioritized recommendations from your audits, your source intelligence, and your content gaps.
### Review pending recommendations
Hikoo lists the recommendations it has generated. Each one is a proposed task waiting for your decision.
### Accept the ones you want
Accept a recommendation to add it to your board as a task. Leave the rest pending.
### Add and reorder manual tasks
You can also add your own manual tasks and reorder tasks to match your priorities.
### Every task shows its evidence
Each generated task links back to what it came from, whether an audit, a source, a content gap, or a citation. Open the link to see the evidence behind the task, so you know why it was proposed.
## Reminders and filters
In settings you can set two reminder times:
* A **daily reminder time** for your tasks in general.
* A separate **high-priority reminder time** for your most urgent tasks.
You can filter the board by **category** and by **status** to focus on one slice of the work.
## Related guides
Score your pages and feed findings into the plan.
Find the third-party sources that win citations.
Close the content gaps behind your AI visibility.
# MCP overview
Source: https://docs.tryhikoo.com/guides/mcp/overview
Use Claude and other AI assistants to explore your Hikoo data through the Model Context Protocol.
Hikoo runs a hosted **Model Context Protocol (MCP) server** so you can ask Claude, or any MCP-compatible client, about your GEO data in plain language. Citations, audits, prompts, competitors, source intelligence, and more, without leaving your chat.
Ask things like:
* "How is my site cited compared to competitors this month?"
* "List the pages that failed criteria in my latest audit."
* "Add a prompt about pricing transparency under the pricing topic."
Your client calls the Hikoo MCP server behind the scenes, which reads and writes data in your Hikoo workspaces.
## How it works
Hikoo runs the server. You do not install anything locally. Point your client at our URL.
A single API key identifies you as a user. It can reach every website in the workspaces you belong to.
Read citations, audits, and analytics. Create or update prompts and competitors. Queue new audits.
The key is scoped to **your user account**, not to a single website. Your client first discovers which websites you have, then passes the chosen `website_id` to each data tool. This replaces the older per-website key model.
## What you get
The server ships **70 tools** across these areas:
* **Citations and sources.** Visibility trends, top platforms, share of voice, top source domains, and citation breakdowns.
* **Sentiment.** Summary and trend of how AI answers portray your brand (available on some plans).
* **Prompts and topics.** Create, update, batch-update, and delete prompts; organize them into topics; read per-prompt stats and citations.
* **Competitors.** Share-of-voice ranking, per-competitor citations, suggestions, and add, update, or remove competitors.
* **Audits.** List past audits, inspect page-level criteria, and queue full-site, single-URL, or selected-page audits.
* **Health.** robots.txt, llms.txt, and sitemap status, plus improvement suggestions.
* **Indexation and pages.** Indexed-page inventory, summaries, common errors, and re-crawls.
* **Content gap.** Suggestions where competitors are cited and you are not.
* **Source intelligence.** AI-cited domains ranked by frequency, so an agent can automate listing claims and outreach.
* **Discovery and subscription.** Who you are, your workspaces and websites, and plan details.
## Who should use it
* **Power users** comfortable in Claude Desktop, Claude Code, or any MCP client.
* **Analysts** who prefer ad-hoc questions over dashboards.
* **Teams** building AI workflows that need programmatic access to their GEO data.
## Next steps
Generate an API key and connect your AI client.
Every tool the MCP exposes, grouped by area.
# Connecting the MCP
Source: https://docs.tryhikoo.com/guides/mcp/setup
Generate an API key and wire up Claude Desktop, Claude Code, or any MCP client to your Hikoo account.
Setup is two steps: create an API key, then plug it into your MCP client.
### Generate an API key
Your key identifies you as a user and can reach every website in the workspaces you belong to. You create it at the account level, not per website.
1. Open **Settings**, then the **API keys** section.
2. Click **Create API key**.
3. Give it a descriptive name such as `Claude Desktop` or `Analyst workflow`, and an optional expiry.
4. **Copy the key immediately.** It starts with `hk_` and is shown only once.
Keys inherit your access. Treat them like passwords, never commit them, and revoke a key if it leaks.
See [API keys](/guides/api-keys) for managing keys and the REST API.
### Point your MCP client at Hikoo
The Hikoo MCP endpoint is:
```
https://mcp.tryhikoo.com/mcp
```
It uses the streamable HTTP transport. Pass your API key in the `Authorization` header.
Edit `claude_desktop_config.json`:
```json theme={null}
{
"mcpServers": {
"hikoo": {
"type": "http",
"url": "https://mcp.tryhikoo.com/mcp",
"headers": {
"Authorization": "Bearer hk_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
}
}
}
}
```
Restart Claude Desktop. The **Hikoo** server appears in the tools menu.
Use `claude mcp add` (or edit `.mcp.json`):
```bash theme={null}
claude mcp add hikoo --transport http \
--url https://mcp.tryhikoo.com/mcp \
--header "Authorization: Bearer hk_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
```
Any client that supports the **streamable HTTP** transport works. Configure:
* **URL**: `https://mcp.tryhikoo.com/mcp`
* **Header**: `Authorization: Bearer hk_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx`
Check your client's documentation for how to supply custom headers.
### Pick a website
Because one key can reach several websites, most questions start by choosing one. Your client handles this for you with the discovery tools:
1. `hk_list_workspaces_with_websites` returns your workspaces and the websites inside each, each with its `has_active_subscription` status.
2. If you have not said which site to use, the client asks.
3. The chosen `website_id` is passed to every data tool.
Data tools return `403` for a website without an active subscription. Discovery tools always work.
### Try it out
Start a new chat and ask something like:
> *"For my main website, what are the top AI platforms citing me this month?"*
The client discovers your site, calls a Hikoo tool, and answers with live data.
## Managing keys
* **List your keys** in **Settings**, then **API keys**.
* **Revoke** a key to deactivate it immediately. Clients using it receive a `401 Unauthorized`.
* **Expiration** is optional. Set it when creating a key if you want automatic rotation.
## Troubleshooting
| Symptom | Likely cause |
| -------------------------------- | ---------------------------------------------------------------------------------------------- |
| `401 Invalid or expired API key` | The key was revoked, expired, or typed incorrectly. Create a fresh one. |
| `403` on a data tool | The chosen website has no active subscription, or you lack access to it. |
| Tools do not appear in Claude | Confirm the client restarted after editing the config, and the transport is `http`, not `sse`. |
See the [tools reference](/guides/mcp/tools) for what you can ask once connected.
# Tools reference
Source: https://docs.tryhikoo.com/guides/mcp/tools
Every tool the Hikoo MCP exposes, grouped by area.
The Hikoo MCP ships **70 tools** across the areas below. You never call them directly. Describe what you want in natural language and your client picks the right one.
Four **discovery** tools take no `website_id`. Every other tool is a **data** tool and requires a `website_id`, which your client obtains from discovery first.
Each tool is labelled so clients know how to treat it:
| Label | Meaning |
| ------------- | --------------------------------------------------- |
| `read-only` | Never changes data. Safe to call freely. |
| `write` | Creates or updates data. |
| `destructive` | Permanently removes data. Review before confirming. |
## Discovery
No `website_id` needed. Start here to find the site you want to work on.
The user your API key is bound to (id, email, name).
Flat list of the workspaces you belong to, with your role in each.
One-shot tree of your workspaces and the websites inside each, with `has_active_subscription` and plan tier. The preferred way to pick a site.
Paginated websites with subscription status.
**Args:** `workspace_id` (optional), `page`, `page_size`
## Website and subscription
Website metadata: url, name, aliases, topics, competitors, language, and subscription status.
**Args:** `website_id`
Full plan detail: tier, status, trial and expiry, tracked AI platforms, feature availability, and every quota. Owner or admin only.
**Args:** `website_id`
## Audits
Audit quota used, limit, and remaining for the current period.
**Args:** `website_id`
Every audit ever run for the website, most recent first, with scores, status, and page counts.
**Args:** `website_id`
One audit: overall score, timing, and the list of audited pages.
**Args:** `website_id`, `audit_id`
Page-level result with every criterion and its pass/fail status, plus the recommendations on each failing criterion. The most detailed audit view.
**Args:** `website_id`, `audit_id`, `page_id`
> *"Why did this page fail in my latest audit? Walk me through each criterion."*
Queue a new full-site audit. Runs asynchronously and counts against your quota.
**Args:** `website_id`
Queue an audit of a single URL. The URL must be on the website's domain.
**Args:** `website_id`, `url`
Queue an audit of a specific list of URLs, capped by your plan.
**Args:** `website_id`, `urls`
Re-run an audit, reusing a prior audit's page set and criteria.
**Args:** `website_id`, `audit_id`
## Citations
Paginated AI citations: the content the assistant produced, its sources, and which competitors were mentioned.
**Args:** `website_id`, `page`, `page_size`, `time_range`, `website_cited`
> *"Show me the last 5 citations where my site was mentioned."*
One citation in full: the assistant's answer, sources, brand mentions with sentiment, the triggering prompt, and competitor ranks.
**Args:** `website_id`, `citation_id`
Top AI platforms by citation count.
**Args:** `website_id`, `time_range`
Share-of-voice trend per platform over time.
**Args:** `website_id`, `time_range`
Top source domains or URLs that appear in AI citations.
**Args:** `website_id`, `time_range`, `group_by` (`domain` or `url`), `limit`
Uncapped, paginated list of every source URL by frequency.
**Args:** `website_id`, `page`, `page_size`, `time_range`, `source_class` (`all`, `mentioned`, `background`)
Counts per match type (brand citation, site mention, background source) with a delta and a sparkline. Capped at 1 year.
**Args:** `website_id`, `time_range`, `tz`
## Sentiment
Available on some plans. Both tools return `403` if sentiment analysis is not on your plan, and cover up to 1 year.
Positive, negative, and neutral counts per brand (your site, competitors, unresolved).
**Args:** `website_id`, `time_range`
Sentiment counts bucketed over time.
**Args:** `website_id`, `time_range`, `bucket` (`day` or `week`)
## Competitors
Competitor ranking (share of voice) against your website.
**Args:** `website_id`, `time_range`
> *"How do I rank against my competitors this quarter?"*
Citation counts per competitor.
**Args:** `website_id`, `time_range`
Share-of-voice chart data for your site plus competitors.
**Args:** `website_id`, `time_range`
Time series of your share of voice versus all competitors.
**Args:** `website_id`, `time_range`
The maximum number of tracked competitors on your plan.
**Args:** `website_id`
Competitors auto-discovered from your citations (available on some plans).
**Args:** `website_id`
Add a tracked competitor. Enforces your competitor limit.
**Args:** `website_id`, `name`, `url`, `aliases`
Update a competitor's name, URL, or aliases.
**Args:** `website_id`, `competitor_id`, `name`, `url`, `aliases`
Hide a suggested competitor.
**Args:** `website_id`, `suggestion_id`
Promote a suggestion to a tracked competitor. Enforces your limit, and is available on some plans.
**Args:** `website_id`, `suggestion_id`, `url`, `logo_url`, `aliases`
Remove a tracked competitor.
**Args:** `website_id`, `competitor_id`
## Content gap
Available on some plans. A limit of 0 means content gap is not on your plan.
Monthly content-gap quota used, limit, and reset date.
**Args:** `website_id`
Counts of gap suggestions by status and priority.
**Args:** `website_id`
Paginated gap suggestions, sorted by similarity (the lowest is the biggest gap).
**Args:** `website_id`, `status`, `min_score`, `max_score`, `limit`, `offset`
Full gap detail: brief, outline, reasoning, the competitor topic, and your closest existing page.
**Args:** `website_id`, `suggestion_id`
## Health
Global website health (AI crawlers allowed, `llms.txt` present, sitemap reachable) plus improvement suggestions.
**Args:** `website_id`, `include_dismissed`
Monthly health-refresh quota used, limit, and reset date.
**Args:** `website_id`
Monthly suggestion-regeneration quota.
**Args:** `website_id`
Paginated indexed pages affected by a health suggestion.
**Args:** `website_id`, `suggestion_id`, `page`, `per_page`
Force a re-check of `robots.txt`, `llms.txt`, and sitemap. Quota-limited.
**Args:** `website_id`
Hide a health suggestion.
**Args:** `website_id`, `suggestion_id`
Un-hide a dismissed health suggestion.
**Args:** `website_id`, `suggestion_id`
Regenerate suggestions from your indexed pages. Quota-limited.
**Args:** `website_id`
## Indexation and pages
Aggregated page stats: total, by category, pass rate, errors and warnings, and last crawl.
**Args:** `website_id`
The latest indexation job status and progress counters.
**Args:** `website_id`
Paginated indexed pages, filterable.
**Args:** `website_id`, `category`, `health_passed`, `has_errors`, `has_warnings`, `language`, `search`, `limit`, `offset`
One indexed page: URL, title, language, category, last crawl, health, and errors.
**Args:** `website_id`, `page_id`
The most common health-check errors grouped by type with counts.
**Args:** `website_id`, `limit`
Trigger a re-crawl of a single URL on the domain. Poll status afterward.
**Args:** `website_id`, `url`
## Prompts
Prompts used versus your plan limit. Call this before bulk-creating prompts.
**Args:** `website_id`
All prompts, grouped by topic, each with its visibility score.
**Args:** `website_id`, `topic_id`
Paginated citations tied to a prompt.
**Args:** `website_id`, `prompt_id`, `page`, `page_size`
Drill-down for a prompt: visibility with a 7-day delta, best rank and platform, coverage, a 30-day sparkline, a per-platform grid, and the latest answers.
**Args:** `website_id`, `prompt_id`, `answers_limit`
Aggregated visibility stats across all of a prompt's citations.
**Args:** `website_id`, `prompt_id`
Create a prompt. Enforces your prompt quota.
**Args:** `website_id`, `content`, `topic_id`, `localizations`
> *"Add a prompt 'best project management tool for startups' under the productivity topic."*
Update one prompt. Only the fields you provide change.
**Args:** `website_id`, `prompt_id`, `content`, `topic_id`, `localizations`, `intent`, `journey_stage`
Apply the same metadata edit to many prompts at once (up to 1500).
**Args:** `website_id`, `prompt_ids`, `topic_id`, `intent`, `journey_stage`
Delete a prompt.
**Args:** `website_id`, `prompt_id`
## Topics
Every topic tracked for the website.
**Args:** `website_id`
Create a topic for organizing prompts.
**Args:** `website_id`, `name`
Rename a topic.
**Args:** `website_id`, `topic_id`, `name`
Delete a topic.
**Args:** `website_id`, `topic_id`
## Search queries
The unique search-query variants AI assistants executed for your site.
**Args:** `website_id`, `page`, `page_size`, `platform`
Query variants triggered by one prompt.
**Args:** `website_id`, `prompt_id`, `page`, `page_size`
Query variants attached to one citation.
**Args:** `website_id`, `citation_id`, `page`, `page_size`
## Source intelligence
Aggregated source counts by `source_type` and by AI platform. Call this first to know what exists.
**Args:** `website_id`
> *"What kinds of sources are citing my space, and on which platforms?"*
Suggested AI-cited sources for outreach, ordered by citation count.
**Args:** `website_id`, `source_type`, `platform`, `search`, `limit`, `offset`
> *"List the top review platforms I should claim a profile on, then draft outreach for each."*
Visibility is capped to your plan's most-cited sources. Upgrading unlocks more sources, not different data.
## Argument conventions
* **`time_range`** accepts: `all_time`, `1y`, `1m`, `1w`, `1d`. Sentiment and breakdown tools cap the range at 1 year.
* **Pagination** comes in a few styles. Most tools use `page` and `page_size` (1-based), content gaps and pages and suggested sources use `limit` and `offset`, and health suggestion pages use `page` and `per_page`.
* **Enums** you may pass:
* Prompt `intent`: `informational`, `comparative`, `transactional`, `navigational`.
* Prompt `journey_stage`: `discovery`, `consideration`, `decision`, `retention`.
* Search-query `platform`: `chatgpt`, `perplexity`, `gemini`, `claude`, `copilot`.
* Source `source_type`: `review_platform`, `media`, `blog`, `directory`, `forum`, `wiki`, `social`.
* Top-sources `group_by`: `domain`, `url`. Sources `source_class`: `all`, `mentioned`, `background`.
* Content-gap `status`: `pending`, `in_progress`, `completed`, `dismissed`.
* **Write and destructive tools** change data in your workspace. Claude Desktop and Claude Code ask for confirmation before running them by default.
# Pages
Source: https://docs.tryhikoo.com/guides/pages
Browse the pages Hikoo has indexed for your site, with category, language, and status for each
Pages is your indexed-page inventory. It is part of the **Analyzer** module. Hikoo discovers your pages from your sitemap, stores them, and enriches each one so you can see the state of your whole site in one table.
For every page Hikoo:
* Auto-categorizes it as home, blog post, product, e-commerce product page, service, FAQ, or landing page.
* Detects its language.
* Detects E-E-A-T signals, covering experience, expertise, authoritativeness, and trustworthiness.
## The pages table
Each row is one indexed page. The table shows:
| Column | Description |
| ----------------------- | --------------------------------------- |
| **URL** | The address of the page. |
| **Category** | The auto-detected page type. |
| **Language** | The detected language. |
| **Status** | Whether the page is `ok` or in `error`. |
| **Last indexed** | When Hikoo last indexed the page. |
| **Errors and warnings** | Any issues detected on the page. |
### Filters
Narrow the table with the available filters:
* **Status** to show only pages that are ok or in error.
* **Category** to focus on one page type.
* **Language** to focus on one language.
* **Search** to find a page by URL.
## Page detail preview
Open a page to see its preview, which shows:
* The **meta description**.
* The **HTTP status**.
* Whether the page is **indexable**.
* A link to its **audit score**, so you can jump to the full per-page breakdown in the [audit](/guides/audit).
## Manual actions
You can manage individual pages by hand:
* **Index a single page** to add it to the inventory.
* **Re-categorize a page** if the auto-detected category is wrong.
* **Delete a page** you do not want tracked.
The number of indexed pages is capped by your plan.
Indexation itself is started and managed from the [Health](/guides/health) page. The Pages table is where you review and manage the results.
## Related guides
Start indexation and check your site-wide AI readiness.
See the per-page AI-readiness score behind each page.
Close content gaps and generate articles from your findings.
# Prompts
Source: https://docs.tryhikoo.com/guides/prompts
Define the natural-language questions Hikoo runs against AI assistants to test your visibility
A **prompt** is a natural-language question that Hikoo runs against AI assistants to see whether your brand gets cited. Your prompts are the tracked query set behind every citation, so writing good ones is the foundation of the whole dashboard.
## The prompts page
The Prompts page shows your prompts as a list, with stat cards across the top for quick health: **All**, **Winning**, **At risk**, **Unclaimed**, and **Average visibility**.
Each prompt row summarizes how you are doing on that question:
* **Visibility score** and **citation rate**
* The **platforms** that cite you on it
* The **competitors** cited alongside you
* The **markets** it is tracked in
* **Sentiment** of the answers
* A **status** of winning, at risk, losing, or unclaimed
## Add a prompt
You can add prompts three ways.
Write a single prompt yourself. Use this when you know the exact question you want to track.
Give Hikoo a set of keywords and let it draft prompts for you. Review and keep the ones that fit.
Drop a CSV into the import dropzone. Hikoo shows a preview, auto-creates any missing topics, and highlights row-level errors before you commit the import.
## Classify prompts
Classification helps you slice your prompts by how buyers actually ask. Each prompt can carry:
* A **search intent**: informational, comparative, transactional, or navigational.
* A **journey stage**: discovery, consideration, decision, or retention.
* A **source keyword** it came from.
These values can be AI-generated, so you can classify a large set without tagging each one by hand.
## Organize and act in bulk
Group prompts by **topic**, by **journey stage**, or by none, then search and filter to find what you need. Select prompts to run bulk actions: **select all**, **delete**, **tag**, **retarget market**, **move to topic**, or **export**.
## Prompt volume
**Prompt volume** estimates how much AI search demand a prompt has, so you can prioritize the questions people actually ask. It is available on some plans.
## Prompts beyond your plan
Your plan sets how many prompt slots each website gets, and a usage bar shows your current use. If you go over that limit, the extra prompts are paused under **Beyond your plan** and can be restored. Adding more slots reactivates them.
Paused prompts stop collecting new citations until you restore them.
## Prompt detail
Open a prompt to see how it has performed:
* **Visibility over the last 30 days**
* Your **best rank** and **coverage**
* A **per-platform breakdown**
* The **latest answers** AI assistants gave
* The **citations** and **competitor citations** tied to the prompt
Write prompts the way real users ask AI assistants. Natural, specific questions match how people actually query, which makes your tracking reflect real demand.
## Related
Group prompts into themes that drive filtering everywhere.
The answers your prompts produce, matched and ranked.
What AI assistants actually search while answering your prompts.
# Search queries
Source: https://docs.tryhikoo.com/guides/search-queries
See the sub-questions AI assistants search on your behalf while answering your prompts
A **search query** is a sub-question that an AI assistant internally issues to a search engine, such as Google or Bing, while it answers one of your prompts. Search queries show you the keywords AI is actually searching for on your behalf, which is often not the same wording you used in the prompt.
## Prompts versus search queries
The difference matters:
* A **prompt** is what you define. It is the question you want to track.
* A **search query** is what the AI actually searched to answer that prompt.
One prompt can spawn many search queries, because the assistant rephrases and breaks your question into several searches. Reading the search queries tells you which keywords are worth winning.
## The search queries page
The page opens with KPIs for **queries tracked**, **topics covered**, and your **top platform**.
Below that, search queries are grouped by topic, with an **Unclustered** group for queries that do not fit a topic yet. You can:
* **Sort** by most generated, most queries, or most recent.
* **Expand** or **collapse** all groups at once.
* **Copy** a query to reuse it elsewhere.
## Query detail
Open a search query to see:
* A **generations over time** timeline showing when it was issued.
* The **platforms** that issued it.
* The **source prompts** it came from.
## Scoped views
You do not have to read every search query at once. You can also view search queries scoped to a single prompt or to a single citation, which keeps you focused on the answer you are investigating.
## Related
The questions you define, which spawn these search queries.
View the search queries behind a single citation.
# Sentiment
Source: https://docs.tryhikoo.com/guides/sentiment
Measure how positively or negatively AI assistants portray your brand and competitors
**Sentiment** measures how positively or negatively AI assistants portray your brand, and your competitors, in their answers. It turns the tone of AI responses into something you can track over time and compare across brands.
## Two views
The summary shows positive, neutral, and negative counts per brand. It includes your site and each of your tracked competitors, so you can see who AI speaks well of and who it does not.
The trend view plots sentiment over time, bucketed by day or by week. Use it to catch a change in tone after a launch, a review, or a competitor's move.
## Sentiment on the citations page
Sentiment is not limited to this page. On the [Citations](/guides/citations) page it appears as both a column and a filter, so you can jump straight to the positive or negative answers behind any change you see here.
## Availability
Sentiment is available on some plans, and covers up to the last year of data.
## Related
Filter citations by sentiment to read the answers behind the trend.
Compare how AI portrays you against each rival.
# Settings
Source: https://docs.tryhikoo.com/guides/settings
Manage your profile, preferences, security, API keys, team, and billing from one modal.
Settings opens as a modal from the user menu. It gathers everything about your account, your workspaces, and your billing in one place. Each area below is a section of that modal.
## Profile
Set your display name. Your email address is shown for reference and cannot be changed here.
## Preferences
Choose how Hikoo looks and reads.
* Theme: Light, Dark, or Auto. Auto follows your operating system's preference.
* Language: English or Français.
Changing the language reloads the app and applies the new locale everywhere.
## Security
Click **Change password** to set a new password. You enter your current password and choose a new one.
If you signed up with Google, password management happens in your Google account rather than here.
## API keys
Create and revoke API keys, and get install instructions for the Hikoo MCP server and the REST API. This is where you generate a key, name it, and copy it once. See [API keys](/guides/api-keys) for the full workflow and safe handling.
## Sites, Members, and Workspace
These sections manage the sites in a workspace, the people on the team, and the workspace's own settings.
* Sites: list the workspace's sites, add a site, manage each site's subscription, and set its language.
* Members: invite teammates, set roles, and manage access.
* Workspace: rename the workspace and access its danger zone.
See [Workspaces](/guides/workspaces) for how roles and per-site subscriptions work.
## Billing
Review your subscription and your billing information. Billing information includes your name, company, address, and tax identifiers such as an EU VAT number or a French SIREN, which are validated automatically. You can also open the Stripe customer portal to update your payment method, view invoices, and change or cancel your plan.
## Related pages
Create keys for the REST API and the MCP server.
Manage sites, members, and roles.
# Source intelligence
Source: https://docs.tryhikoo.com/guides/source-intelligence
Find the third-party sources AI assistants pull from and see where your brand is absent
Source intelligence finds the third-party sources that AI assistants read when they answer, then shows you where your brand is present and where it is missing. It answers a single practical question. Where do I need to be listed or mentioned to earn AI citations?
The sources it tracks include review platforms, media, blogs, directories, forums, and social. When an AI assistant leans on one of these to build an answer, Source intelligence records it and tells you whether you were part of that source or not.
Source intelligence is available on some plans, and the number of sources you can see depends on your plan.
## Two ways to look at sources
You can switch between two views:
* **By domain** groups everything by the site it lives on.
* **By article** breaks it down to the individual page or post.
At the top, hero KPIs give you the headline picture: your **visibility share**, your **engine coverage**, and your **top publishers**. A set of cards adds context: **total sources**, **gaps detected**, **average GEO score**, and whether **you are present**.
## Families and relation
Every source is sorted into a family and a relation, so you can scan the landscape quickly.
**Buckets (source families):**
| Family | What it covers |
| ---------- | ----------------------------------- |
| Affiliate | Pages that earn from referrals. |
| Comparison | Head to head and versus pages. |
| Review | Review platforms and rating sites. |
| Media | News outlets and publications. |
| Community | Forums and discussion sites. |
| Video | Video platforms. |
| Reference | Encyclopedic and reference sources. |
**Relation to you:**
| Relation | Meaning |
| ----------- | ---------------------------------- |
| Present | You already appear on this source. |
| Lost ground | You were present and slipped. |
| Contested | You and competitors both appear. |
| Owned | You control this source. |
| Open | Nobody has claimed it yet. |
## Filtering
Narrow the list until it shows only what matters right now:
* **Search** by keyword.
* **Engine or platform** to focus on one AI assistant.
* **Family** and **bucket** to focus on a source type.
* **Presence** to show sources where you are present or absent.
* **Relation** to focus on contested, owned, open, and more.
A **gaps-only** toggle hides everything except the sources where you are missing. You can also filter to the sources fueling a specific competitor, so you can see exactly what is feeding their citations.
## Opportunities, matrix, and the full list
Three sections help you move from insight to action:
* **Top opportunities** surfaces the highest-value sources to pursue.
* A **competitor matrix** compares who appears where.
* An **All sources** table lists everything in one place.
The All sources table carries these columns:
| Column | Description |
| ------------- | -------------------------------------- |
| **Source** | The page or publication. |
| **Domain** | The site it lives on. |
| **Type** | The source family. |
| **Used by** | Which brands this source cites. |
| **Frequency** | How often AI assistants use it. |
| **Lift** | The projected gain from being present. |
## Source dossier
Open any source to see its dossier, a full brief on why it matters and what to do.
* A **GEO score** out of 100.
* A projected **lift** if you get listed.
* **Who to reach out to** to earn a mention.
* **Why you are absent** today.
* **Citations by platform** for this source.
* The **discovery origins**, the queries that surfaced the source in the first place.
* **Citation history** over time.
* **Get listed** and **Visit site** actions.
* An **I'm on it** tracker so you can mark a source as in progress.
You can also **dismiss** a source you do not want to pursue, and **restore** it later if you change your mind.
## Presence detection
Presence detection tells you whether you are already mentioned on a source. When Hikoo thinks you may already be there, you can **verify** the mention, then **confirm** it or **dismiss** it. This keeps your gap list honest, so you only chase sources where you are truly absent.
Non-actionable domains are skipped. Sources such as Wikipedia, `schema.org`, and large marketplaces are excluded because you cannot simply get yourself listed on them.
## Running a fresh scan
Sources refresh automatically over time. When you want an update sooner, run one yourself.
### Open the dialog
Click **New analysis**.
### Choose your scope
Pick the **platforms**, **keywords**, and **competitors** to include in the scan.
### Run it
Start the scan. New sources appear as Hikoo discovers them.
## Where to go next
See how source presence maps to your competitive standing.
Pair source gaps with the content gaps you can fill.
Trace a source back to the answers it appears in.
# Topics
Source: https://docs.tryhikoo.com/guides/topics
Group prompts into themes that drive filtering across the dashboard
A **topic** (also called a subject) groups your prompts into themes. Topics keep a large prompt set organized, and they act as one of the shared filter dimensions across the dashboard, so the way you organize topics shapes how you read every other view.
## Manage topics
From the Topics page you can:
* **Add** a new topic.
* **Rename** an existing topic.
* **Delete** a topic you no longer need.
* **Search** to find a topic quickly.
## How topics flow through the dashboard
On the [Prompts](/guides/prompts) page, prompts can be shown grouped by topic. Topic is also one of the dimensions in the shared filter bar, so it carries into citations, competitors, and search queries as you navigate. Clear topics make every one of those views easier to slice.
Keep topics aligned with how buyers think about your space, not with your internal team structure. When a topic matches a real buyer theme, filtering by it answers a real question about your visibility.
## Related
Add prompts and group them by topic.
See how the shared filter bar applies topics everywhere.
# Websites
Source: https://docs.tryhikoo.com/guides/websites
Track multiple websites from one account, each with its own prompts, citations, competitors, and audits.
A website is the unit everything in Hikoo is scoped to. Each website keeps its own prompts, citations, competitors, and audit data, so you can track several brands or domains from a single account without mixing their results.
## Add a website
Open the site switcher in the sidebar and choose to add a site. A short wizard walks you through the basics.
Provide the website URL and a name you will recognize in the switcher.
Choose a target market and localization so Hikoo tracks the right country and language, then add a short description of what the site is about.
On creation Hikoo analyzes and crawls the site to seed its metadata, topics, and a set of starter prompts. You can refine all of these afterward.
The starter prompts are a starting point, not a final list. Review them and add your own on the [Prompts](/guides/prompts) page.
## Edit a website's details
Each site has details you can edit at any time:
| Detail | What it controls |
| -------------------------- | ---------------------------------------------------------------------------------- |
| URL | The primary domain Hikoo tracks. |
| Name | The label shown in the site switcher and across the app. |
| Description | A short summary of the site, used to guide analysis. |
| Topics | The themes that group your prompts. |
| Brand aliases | Alternative names your brand is known by, so Hikoo counts citations that use them. |
| URL aliases | Additional URLs that should count as your site. |
| Competitors | The brands you benchmark against. See [Competitors](/guides/competitors). |
| Language and localizations | The country and language markets you track. |
Brand aliases matter for accurate citation counts. If AI assistants refer to your brand by a shortened name or a former name, add it here so those mentions are attributed to you.
## Switch between sites
Use the site switcher in the sidebar to change the active site. Sites are grouped by workspace, so you can see which team each one belongs to. From the switcher you can also:
* Add a site.
* Add a workspace.
* Search for a site by name.
* Open management for your sites and workspaces.
The active site drives every page in the dashboard, so the citations, prompts, and audits you see always belong to the site selected here.
## Website limits
Your plan sets how many websites you can track. As you approach that number you will see clear states for tracked sites, sites approaching the limit, and the limit being reached. When you reach the limit you will need to remove a site or move to a plan that allows more.
## Delete a website
You can delete a website from its management view.
Deleting a website permanently removes all of its data, including citations, prompts, audits, and competitors. This cannot be undone.
## Related pages
Group sites and teammates under a workspace and its subscription.
Manage the tracked query set for each site.
Benchmark a site's visibility against its rivals.
# Workspaces
Source: https://docs.tryhikoo.com/guides/workspaces
Organize websites and teammates under a workspace with its own roles and subscription.
A workspace is a team container. It holds websites and a subscription, and it is where you invite teammates and manage their access. You can belong to several workspaces at once, and the site switcher groups your websites by the workspace they live in.
## Roles
Every member of a workspace has a role that sets what they can do.
| Role | What they can do |
| ------ | ----------------------------------------------------------------------- |
| Owner | Full control, including the plan, settings, and transferring ownership. |
| Admin | Manage the plan, settings, sites, and members. |
| Member | Work with sites and data, without managing the plan. |
| Viewer | Read-only access. |
A Viewer can see the workspace but cannot make changes. Write actions are blocked, with a tooltip explaining why and a toast if they try. Owners and Admins are the roles that manage the plan and workspace settings.
## Manage members
From the Members section you can:
* Invite a teammate by email and assign a role.
* See pending invitations and revoke one that has not been accepted.
* Change an existing member's role.
* Remove a member from the workspace.
* Transfer ownership to another member.
Transferring ownership hands the Owner role to another member. Do this when the person responsible for the plan and workspace changes.
## Where workspaces are managed
Workspaces are managed from the Settings modal, which opens from the user menu. The relevant sections are:
Rename the workspace. A danger zone lets you delete the workspace or leave it.
List the sites in the workspace, add a site, manage each site's subscription, and set its language.
Invite, review, and manage the people in the workspace and their roles.
Review per-site subscription status, open the Stripe portal, and view invoices. A summary shows how many sites are subscribed out of the total.
Billing in a workspace is per site. The Billing section shows a "subscribed sites X of Y" summary so you can see coverage at a glance.
## Related pages
Where the workspace, members, and sites sections live.
Add and edit the sites a workspace tracks.
# Welcome to Hikoo
Source: https://docs.tryhikoo.com/index
Track and improve how AI assistants cite your website.
Hikoo is a GEO and AEO platform (Generative and Answer Engine Optimization). It shows you how AI assistants like ChatGPT, Perplexity, Claude, Gemini, and Google AI Overviews cite your website when they answer real questions, benchmarks that visibility against your competitors, and helps you close the gaps.
## Get started
Add your first website and start tracking AI citations in minutes.
The core concepts: prompts, citations, share of voice, and AI readiness.
## Spotlight: see your AI visibility
Every time an AI platform references your content, with full context.
The questions Hikoo runs against AI assistants to track your brand.
The searches AI assistants run behind the scenes to answer your prompts.
How positively AI assistants portray your brand and competitors.
## Battlemap: outrank your competitors
Benchmark your citation share and reclaim lost ground.
Find the third-party sources AI cites, and where you are missing.
## Elevate: create what earns citations
Find the content gaps where competitors get cited and you do not, with a brief for each one.
## Analyzer: fix your AI readiness
Score pages against AI-readiness criteria and track improvements.
Check robots.txt, llms.txt, sitemap, and AI crawler access.
Your indexed page inventory with categories and errors.
Turn findings into a prioritized GEO task board.
## Connect your AI assistant
Query your Hikoo data from Claude, Claude Code, or any MCP client in plain language.
# Quickstart
Source: https://docs.tryhikoo.com/quickstart
Set up your Hikoo account and see your first AI-visibility snapshot.
## Create your account
### Sign up
Go to the [Hikoo app](https://app.tryhikoo.com) and create an account with your email and password, or sign in with Google. New accounts go straight into onboarding.
### Add your website
Enter your website URL and name, and choose the market you want to track (worldwide or a specific country). Hikoo analyzes and crawls the site to seed its metadata, topics, and starter prompts.
### Confirm your profile
Review the description, industry, and brand identity Hikoo detected, and adjust the topics it will track. Accurate topics make every later view easier to slice.
### Add competitors
Add the competitors you want to benchmark against. You can skip this and add them later from the [Competitors](/guides/competitors) page.
### Select prompts
Hikoo generates a set of prompts (the questions it will ask AI assistants). Select the ones that match how your audience searches, regenerate for more ideas, and confirm.
### See your snapshot
Hikoo runs a first pass and shows a live snapshot of your AI visibility, so you get value before you pay.
### Choose a plan
Start a 7-day free trial with no credit card, pick the Free plan, or choose a paid plan and check out. Plans differ by websites, prompts, AI-response runs, audits, indexed pages, and which features they include.
## What happens next
Once your website is set up, Hikoo works in the background:
* **Indexation.** Your site is crawled to discover and categorize its pages. See [Pages](/guides/pages).
* **Citation collection.** Hikoo queries the AI assistants for your prompts every day and records when they cite you. See [Citations](/guides/citations).
* **Health checks.** Your site is evaluated for AI readiness, including `robots.txt` crawler access, `llms.txt`, and sitemap presence. See [Website health](/guides/health).
Data fills in on your dashboard as results arrive.
## Next steps
See your visibility metrics at a glance.
Learn how prompts, citations, and share of voice fit together.
Add, import, or AI-generate the questions Hikoo tracks.
Ask Claude about your Hikoo data in plain language.