Skip to main content
Glama

Server Details

How ChatGPT, Gemini, Perplexity and Google AI name your brand, cite sources and rank rivals.

If you are the author of this connector, you can claim ownership. Claimed connector authors can inspect health checks, view analytics, and manage their listing.
Status
Healthy
Uptime
100.0% over the last 24 hours
Last Tested
Transport
Streamable HTTP · MCP 2025-11-25
URL
Repository
howseen-ai/howseen-mcp
GitHub Stars
0
ChatGPT
–
Claude
–

TDQS

A4.1/5.0

Scored across 12 tools

Disambiguation4/5

The set draws a clean read/write split (get_* read-only vs manage_*/create_*/publish_*/update_* mutating) and each tool targets a distinct domain object: visibility, prompts, content, traffic, actions, settings. The only soft boundaries are between get_content (articles/site pages), manage_articles (writes but 'never publishes') and publish_article (publish/index/link), which an agent must read carefully; the heavy use of view/operation sub-modes inside each tool adds intra-tool choice but not tool-to-tool confusion.

Naming Consistency5/5

Every name follows a consistent snake_case verb_noun pattern with a small, predictable verb vocabulary (get_, manage_, create_, publish_, update_). The verbs also encode the read vs write distinction, so the naming itself reinforces intent.

Tool Count5/5

Twelve tools is well within the sweet spot and each one owns a real slice of the product (visibility, prompts, content, traffic, actions, tracking, publishing, settings, sharing). Nothing looks like a redundant wrapper; multi-mode tools absorb what would otherwise be dozens of endpoints.

Completeness4/5

The lifecycle is broadly covered: read analytics, manage tracking and competitors, draft/edit/delete/restore articles, publish and request indexing, manage actions, update profile and brands, and share reports. Minor gaps exist on the share-report side (create only — no list, revoke or update) and connections are viewable but not modifiable, though those are plausibly handled in the app UI.

Available Tools

12 tools
create_share_reportCreate a share reportAInspect

Create a public, frozen report link (visibility, competitors, top questions, recommendations) with the agency branding. Without confirm: true NOTHING changes: you get a preview. Show it to the user and call again with confirm: true only after they approve it.

Inputs: to, from, lang, brand, +2 more.

get_accountAccount, brands and settingsA
Read-only
Inspect

Your Howseen account. Call view brands first when the account may track several brands. view: brands = the tracked brands (id, name, domain); usage = plan, quotas used and left, next scan, what this key may do; settings = brand profile, competitors, writing rules, Autopilot; connections = true state of every connection (CMS, feed, Search Console, GA4, Bing). Read-only.

Inputs: view, brand.

get_actionsWhat to do nextA
Read-only
Inspect

What to do to get cited by AI, ranked like the dashboard. view: next = THE next step, with its steps and how to close it; plan = the full prioritized action plan; easiest_wins = sites cited in the answers the brand loses: where to get listed; site_audit = technical audit with paste-ready fixes (llms.txt, robots.txt, JSON-LD); refresh true re-crawls; pruning = quarterly review: pages to improve, merge or remove. Read-only.

Inputs: to, from, lang, view, +5 more.

get_contentArticles and content planA
Read-only
Inspect

Articles Howseen writes for the brand and the brand's own pages. view: plan = content calendar, next article to write and to publish, stale facts; articles = drafts, scheduled and published articles; performance = indexing status, impressions and clicks per article; site_pages = the brand's own website pages. Read-only.

Inputs: to, from, view, brand, +2 more.

get_promptsBuyer questions and AI answersA
Read-only
Inspect

The buyer questions (prompts) tracked for the brand and the real AI answers to them. view: tracked = tracked questions with visibility and who is named; gaps = questions where a competitor is named and the brand is not; answers = real AI answer excerpts, filter with query / prompt; fanout = web searches the AI ran behind each question; suggestions = questions worth tracking (adds nothing, 20 calls a day). Read-only.

Inputs: to, from, view, brand, +6 more.

get_trafficTraffic and Google dataA
Read-only
Inspect

Traffic of the brand's site from the connected Google accounts. view: ai_referrals = visits and revenue AI assistants send; search_console = clicks, impressions, queries, pages and opportunities; analytics = GA4 sessions, sources and pages (account-wide). Read-only.

Inputs: days, view, brand.

get_visibilityAI visibilityA
Read-only
Inspect

How often ChatGPT, Gemini, Perplexity, Google AI Overviews and AI Mode name the brand. view: overview = visibility %, share of voice, position, per engine, trend; competitors = the brand against its competitors, per engine; changes = what moved since the previous scan (or since a date) and why; sentiment = tone of the answers naming the brand, negative questions; sources = domains and URLs the AI engines cite most; wins = published articles now cited by AI; alerts = drops, competitors overtaking, new competitors (30 days). Read-only.

Inputs: to, from, view, brand, +4 more.

manage_actionsUpdate the action planA
Idempotent
Inspect

Close or dismiss an action-plan item (action_id from get_actions). operation: done = mark it done; dismiss = dismiss it (reversible in the app). Without confirm: true NOTHING changes: you get a preview. Show it to the user and call again with confirm: true only after they approve it.

Inputs: action_id (required), operation (required), brand, confirm.

manage_articlesWrite and manage articlesA
Destructive
Inspect

Write, edit, rewrite, delete or restore an article. Never publishes. operation: draft = write a draft for a question, or for a planned article_id; edit = change the title, meta or exact words of a written article (title, meta, replacements); refresh = rewrite a written article with the current brand profile; delete = move to Trash (restorable); restore = bring back from Trash. Without confirm: true NOTHING changes: you get a preview. Show it to the user and call again with confirm: true only after they approve it.

Inputs: operation (required), meta, brand, title, +5 more.

manage_trackingChange what is trackedA
Destructive
Inspect

Change the tracked questions and competitors, or scan now. Plan quotas apply. operation: add_prompts = track new questions (prompts, topic); archive_prompts = stop scanning, keep history (prompts); remove_prompts = remove from the list (prompts); add_competitor = track a competitor by name or URL (competitor); remove_competitor = stop tracking a competitor (competitor); run_scan = scan the questions on the AI engines now (plan cadence). Without confirm: true NOTHING changes: you get a preview. Show it to the user and call again with confirm: true only after they approve it.

Inputs: operation (required), brand, topic, confirm, +2 more.

publish_articlePublish an articleA
Destructive
Inspect

Put a written article online on the brand's site and get it indexed. operation: publish = publish now on the connected CMS or site, or schedule with publish_at; request_indexing = submit a published article to Google, Bing or IndexNow; inbound_link = link a page of the site to the new article (mode apply, revert or verify). Without confirm: true NOTHING changes: you get a preview. Show it to the user and call again with confirm: true only after they approve it.

Inputs: article_id (required), url, mode, brand, +5 more.

update_settingsUpdate brand settingsA
Destructive
Inspect

Change the brand's profile, writing rules or Autopilot, or add / remove a brand. operation: add_brand = add a brand within the plan's sites (url, name); remove_brand = PERMANENTLY delete a brand and its data (brand, confirm_name); instructions = set or append writing rules, binding on every article; autopilot = turn Autopilot on or off (on); profile = offer, category, about, language, countries, served area. Without confirm: true NOTHING changes: you get a preview. Show it to the user and call again with confirm: true only after they approve it.

Inputs: operation (required), on, url, mode, +12 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    Try in Browser

    Glama MCP Gateway

    Add one secure layer between your agents and this server.