Anry
Server Details
Anry is a read-only MCP server that ties every ad click to the visit and the order, puts a price on where an online store loses money, and gives the fix to the store's own AI agent.
The last two health checks of this connector failed. If this server requires authentication, missing or expired test credentials may be the cause. A test profile lets Glama authenticate for health checks and discover tools; it is separate from your personal connections.
If you are the author, claim ownership, then add or update a test profile under Admin → Test Profile.
- Status
- Unhealthy
- Uptime
- 0.0% over the last 24 hours
- OAuth
- Works in Glama
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- ChatGPT
- –
- Claude
- Anry.io
TDQS
Scored across 23 tools
Most tools have a crisp resource+action identity, and the dense clusters (sessions: list_sessions → get_session → get_session_replay → get_session_frames; reports: list_opportunities vs list_questions vs list_reports vs get_report) are explicitly differentiated by each description's suggested call order. Residual friction remains between analytics tools that all return site metrics over a window (get_analytics_overview, get_top_pages, get_traffic_sources, get_revenue_leaks) and between get_paid_waste_summary and get_campaign_spend, where an agent must read carefully to pick correctly.
Nearly every name is snake_case verb_noun (get_*, list_*, record_diary_entry), with the get_/list_ pair used consistently for single-item vs collection access (get_report/list_reports, get_playbook/list_playbooks, get_session/list_sessions). The one write tool, record_diary_entry, follows the same pattern, so there are no convention breaks.
23 tools is above the comfortable 3-15 band, but the surface spans several legitimately different domains (site analytics, paid ads, session replay, reports/diary, playbooks), so most tools earn their place. The session-inspection cluster (list_sessions, list_notable_sessions, get_session, get_session_replay, get_session_frames) is the main area where a smaller surface could plausibly work.
The investigative lifecycle is well covered end to end: discover sites, check data health, pull aggregate and per-page/per-source metrics, surface opportunities and reports, then drill into individual sessions and visual frames, with a diary write for recording findings. Gaps are minor — no text/keyword search across reports or issues, no explicit comparison/segment tool, and no way to configure goals or ad connections (read-only by design).
Available Tools
23 toolsget_analytics_overviewGet analytics overviewARead-onlyIdempotentInspect
One-call site overview for a date window (default last 30 days, max 31): visitors, page views, bounce rate, average session time, revenue, and conversion rate, plus the top 5 pages and top 5 traffic sources. Start here for questions like 'how is the site doing'.
Inputs: site_id (required), end_date, start_date.
get_campaign_spendGet campaign spendARead-onlyIdempotentInspect
Paid ads per campaign, or per ad with grain=ad, for a site over a date window (default last 30 days, at most 90). Each row has what the platform reports (spend, impressions, clicks, platform conversions) next to what Anry attributed on the site (revenue, conversions, sessions, ROAS), in the site's currency, ordered by spend. Use to connect ad investment with on-site outcome, find which ads or campaigns return revenue, and spot spend with no attributed visit. Narrow with platform or campaign_id when the result is truncated.
Inputs: site_id (required), grain, limit, end_date, +3 more.
get_heatmap_summaryGet heatmap summaryARead-onlyIdempotentInspect
How visitors actually interact with one page (max 31-day window): scroll depth reached by 75/50/25 percent of sessions, plus per-element click intensity (very_high to low vs the expected uniform share) with click shares. Use to understand attention and dead zones on a page before proposing UI changes.
Inputs: site_id (required), pathname (required), end_date, start_date, +1 more.
get_issueGet issue evidenceARead-onlyIdempotentInspect
Full evidence bundle for one issue: metrics, impact, root cause, traffic source and device breakdowns, and user-path analysis. This is the same context Anry's own suggestion pipeline reasons over, so you can derive recommendations yourself or compare with the suggestion block in the subject's report (get_report).
Inputs: issue_id (required).
get_paid_waste_summaryGet sub-floor paid wasteARead-onlyIdempotentInspect
Ad spend the site is losing against its target ROAS on ads too small to have produced a card, summed across the account: the total, how many ads it came from, and the biggest single one. Use when asked where paid money is leaking beyond the flagged ads. Carded findings come from list_opportunities and get_report instead.
Inputs: site_id (required), end_date, start_date.
get_playbookGet playbookARead-onlyIdempotentInspect
One playbook's full Markdown, by name from list_playbooks: a methodology to follow step by step, a reference note, or the operating principles that frame every Anry task.
Inputs: name (required).
get_reportGet reportARead-onlyIdempotentInspect
The full text of one report: a dated, immutable snapshot of what Anry knew about that subject on the day it was written, with every number frozen inline. This is byte for byte the document the team sees, so quoting it can never drift from what they read. The state in the document's header is frozen at write time, not live: check the current state with the list tools before calling a finding open, and cite an archived subject's older snapshots in the past tense. snapshot is 1-based and defaults to the latest. Quote monetary values verbatim.
Inputs: report_id (required), snapshot.
get_revenue_leaksGet revenue leaksARead-onlyIdempotentInspect
The flagship money call: the site's deduplicated revenue (or users) at risk per week and month, plus the top 5 active leaks ranked by weekly potential loss with impact levels. Use for 'where are we losing money' and to pick what to fix first. Quote monetary values verbatim.
Inputs: site_id (required).
get_sessionGet session journeyARead-onlyIdempotentInspect
The ordered in-session journey for one session id (from list_sessions): the page path plus the interaction timeline (page views, clicks, custom events, purchases), oldest event first, with a dashboard replay link. Complements list_sessions with what happened inside a single session. Returns event metadata only, never replay/DOM content.
Inputs: site_id (required), session_id (required).
get_session_framesGet session framesARead-onlyIdempotentInspect
What one session's pages looked like: images drawn from the recording at the moments get_session_replay names (each page, repeated clicks, the browser translation, clicks, field focus, scroll stops, idle starts), with the pointer marked. Without at it returns one contact sheet of up to 12 captioned thumbnails plus the index of every drawn moment; with at (offsets in seconds, as get_session_replay gives them) the full frames nearest those offsets. Use it when the question is visual (what was on screen, covered, below the fold) after get_session_replay. A session not drawn yet answers status rendering: retry after retry_after_seconds.
Inputs: site_id (required), session_id (required), at, max_frames.
get_session_replayGet session replay factsARead-onlyIdempotentInspect
The recording of one session (a session id from list_sessions or get_session) as facts, never as content: a timeline of pages, pointer dwell over elements, clicks with the element clicked, field focus and typed lengths, text selection, scroll depth and idle gaps, plus signals (repeated clicks, browser machine-translation of the page, a per-field form summary, long hovers on buttons and short links, back-and-forth between pages, how the session ended, and seconds, scroll depth and longest idle per page). Call it after get_session when the page journey alone does not explain the behaviour. Not available for reduced-mode sessions or before the recording is stored.
Inputs: site_id (required), session_id (required).
get_site_statusGet site statusARead-onlyIdempotentInspect
Whether a site is set up and sending data, as facts with timestamps: when the tracker first sent an event and whether it sent any in the last 24 hours, when the first purchase event arrived, onboarding, the declared platform, the primary goal, each ad platform connection (status, last sync, error, and over the last 7 days its clicks next to the visits that carried its campaign or ad id) and the store integrations. Call it before reading analytics on a site that looks empty, whenever asked whether something is connected or working, and when paid spend shows no attributed visits. It reports no score and no verdict.
Inputs: site_id (required).
get_top_pagesGet top pagesARead-onlyIdempotentInspect
Per-page metrics for a site over a date window (max 12 months): visitors, sessions, revenue, page views, bounce/exit/conversion rates, ordered by revenue. Use for 'which pages matter' and page-level triage.
Inputs: site_id (required), limit, end_date, start_date.
get_traffic_sourcesGet traffic sourcesARead-onlyIdempotentInspect
Traffic sources for a site over a date window (max 31 days): visitors, conversion rate, revenue, and conversions per source. Use when the question is about where visitors or revenue come from. Pass utm_campaign (or utm_source, utm_medium) to count one campaign's visits by source.
Inputs: site_id (required), limit, end_date, start_date, +3 more.
list_diary_entriesList site diary entriesARead-onlyIdempotentInspect
The site diary for a window, newest first: what the team and its agents recorded as changed on the site, observed, or withdrawn, each with its subject, date, author and the client that wrote it, its evidence, and whether it was later retracted. Read it before diagnosing a page, a campaign or a move in the numbers: a change inside the window is the first explanation to test, and a fix already shipped is not proposed again. Entries are declared by people, not observed by Anry. Defaults to the last 90 days; at most 365.
Inputs: site_id (required), to, from, kind, +2 more.
list_notable_sessionsList notable sessionsARead-onlyIdempotentInspect
Which sessions of a window are worth opening, and how often each behaviour occurred across all of them. Anry reads every recording once and ranks; you write the story. Returns a summary (per behaviour: sessions, share, device split, top pages, the previous window of the same length) and four groups: friction (repeated clicks, a field left empty, browser translation, a long idle or a quiet exit on a key page, a long hover on a button or short link, back-and-forth between two pages, dead-end exploration), abandon (reached cart, checkout or the goal page without converting), paid (paid traffic that left within seconds or never scrolled) and converter (for contrast). Each session carries the reasons it was picked, with offsets in seconds for get_session_replay and get_session_frames, and whether its frames are already drawn. Defaults to the site's previous day; start here before opening recordings one by one.
Inputs: site_id (required), group, limit, end_date, +2 more.
list_opportunitiesList opportunitiesARead-onlyIdempotentInspect
What Anry is watching on this site right now: the carded findings the team sees on their own dashboard, highest priority first, each with its subject, stream, live state, estimated impact, and a report id. This is the curated view and stays small by design, so start here for 'what matters now' and call get_report on a row for the full document. Pass since (ISO 8601) to poll for cards that appeared after a moment you have already seen. Quote monetary values verbatim.
Inputs: site_id (required), since.
list_playbooksList playbooksARead-onlyIdempotentInspect
Lists Anry's playbooks: the operating principles, the invocable methodologies (diagnose conversion opportunities, audit paid ads efficiency, prioritize a CRO backlog) and the reference notes, each with when to use it. Call it at the start of any task about a store's revenue, conversions, paid ads or setup, then load the one that fits with get_playbook.
Inputs: no named parameters.
list_questionsList questionsARead-onlyIdempotentInspect
The site's question watchlist: the questions members asked Anry to keep answering, with each one's rhythm, answer status, and a pointer to its latest answer. Call get_report with a row's report_id for the answer document. On a young site this list fills before list_opportunities does, because the watchlist starts with the team's own scheduled questions.
Inputs: site_id (required).
list_reportsList reportsARead-onlyIdempotentInspect
The full report ledger for a site, newest activity first: every report series it ever had, including archived subjects and cards that aged out of the dashboard. This is the investigation and history view; for what matters now, use list_opportunities and list_questions instead. Filter by state, type, or stream, and page by passing next_cursor back verbatim as cursor. The weekly Brief journal is type brief and belongs to no stream, so stream-filtered calls never return it.
Inputs: site_id (required), type, limit, state, +2 more.
list_sessionsList sessionsARead-onlyIdempotentInspect
Screen recordings (session replays) for a site over a date window (max 31 days), newest first. Every session listed has a recording, and its replay_url opens it in the player: use this tool whenever the user asks for replays or recordings to watch. Each row carries the entry page, duration, landing utm tags and source. Filter by entry_pathname (where the session started), pathname (a page visited at any point), min_duration_seconds, source (the name get_traffic_sources returns), utm tag, or campaign_id / ad_id (the ids get_campaign_spend returns) to see the sessions an ad brought. Meta ads: source is the app the visitor clicked in (Facebook, Instagram or Threads), while utm_source keeps what the ad URL carried, often facebook for every placement, so filter on source=Instagram for an Instagram placement; sessions recorded before this attribution existed show every Meta placement as Facebook. Paginate with the returned next_cursor (pass cursor and cursor_id back verbatim) to collect every match. Each session_id opens with get_session (the page journey), get_session_replay (what the visitor did, as facts) and, where available, get_session_frames (what the pages looked like).
Inputs: site_id (required), ad_id, limit, cursor, +12 more.
list_sitesList sitesARead-onlyIdempotentInspect
Lists the Anry sites the authenticated user is a member of. Call this first: every other tool takes one of these site ids as site_id.
Inputs: no named parameters.
record_diary_entryRecord a site diary entryAInspect
Writes one dated entry to the site diary, the team's shared record of what changed on the site and what was learned, read by every member, every connected agent and Anry's reports. This is the only tool that writes, and it writes nothing but the diary: no setting, campaign, goal or report changes. kind change: something shipped or reverted on the subject. kind observation: a dated fact learned (a test on a phone, a provider's behaviour, a decision). kind retraction: withdraws an earlier entry (retracts_entry_id); nothing is ever edited or deleted. Record what happened, never an intention or a to-do, one entry per fact at the end of an investigation, on the subject you investigated, with the reports and sessions it rests on as evidence. Never put personal data (names, emails, order numbers) in the text. Pass an idempotency_key so a retry never writes twice.
Inputs: kind (required), title (required), site_id (required), body, +5 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceRemote MCP server that lets AI assistants read tracked competitive intelligence — competitor news, reviews, website changes, social posts and ads — and act on it by managing projects, competitors, battle cards, reports and canvases. Every answer carries a source and a date, and write actions only publish when explicitly shared.MIT
- AlicenseCqualityCmaintenanceCompetitor Monitor AI - MCP server providing AI-powered tools and automation by MEOK AI Labs112 npm46 PyPIMIT
- AlicenseAqualityDmaintenanceEnables brand visibility monitoring across major AI platforms like ChatGPT, Claude, Gemini, and Perplexity. It allows users to track visibility scores, analyze competitor data, and receive actionable insights to improve AI-generated brand recommendations.1611 npm1MIT

industrylens-mcpofficial
AlicenseNot gradedqualityCmaintenanceBrowse IndustryLens's published competitive-intelligence reports and head-to-head competitor comparisons from any AI agent — real, source-backed data.MIT
Glama MCP Gateway
Add one secure layer between your agents and this server.