MCP HeartbeatTHE MCP OBSERVATORY
Public workspace

Explore

Evidence for your next connection.

Independent measurements of remote MCP servers. Every observation has a timestamp.

Read-only probes

MCP server · Code & DevOps

ExtensionDash

Research extension listings and search results; track installs, keyword rankings, and competitors.

Checks passed
11 / 22

3 failed · 8 not measured

Median response
603 ms

Median of successful probes over 7 days

30-day uptime
Not measured

Share of countable probes over 30 days, including daily rollups. Auth challenges are excluded.

Discovered tools
16

Tool definitions returned by this server

Independent monitoring

Health & checks

Status history

Responded

The server completed its last measured handshake. This is an observation, not a guarantee.

Last checked
Check summary by test group
Test areaPassedFailedUnmeasured
ConnectionCan a client connect and discover tools?
500
Reliability & speedDoes it respond consistently and quickly?
004
Access & authenticationWhat is required to use this server?
102
Tool qualityAre the tool definitions usable by an agent?
310
Publisher identityCan the publisher be traced to its source?
222

Unmeasured means there is not enough evidence to decide. How these checks work

Inspect individual checks 22 recorded

Core

  • Pass
    Reachable

    Latest probe returned HTTP 200 in 603 ms.

  • Pass
    Valid handshake

    protocolVersion 2025-11-25, serverInfo.name extensions-dashboard.

  • Pass
    Lists tools

    tools/list returned 16 tools.

  • Pass
    Current protocol

    Negotiated 2025-11-25. The bar is 2025-06-18 or newer.

  • Pass
    HTTPS

    The endpoint URL is https.

Performance

  • Not measured
    Fast (p50)

    Median latency 603 ms over 1 samples in 7 days. The bar is 1000 ms. Need at least 5 samples.

  • Not measured
    Consistent (p95)

    95th percentile 603 ms over 1 samples in 7 days. The bar is 3000 ms. Need at least 20 samples.

  • Not measured
    Reliable (uptime)

    1 of 1 attempts returned a tool list in 30 days (100%). The bar is 99%. Need at least 30 samples.

  • Not measured
    Stable tool list

    Fewer than two successful probes in 30 days, so tool-list stability is not measured.

Access

  • Pass
    Open access

    tools/list succeeded with no credential, and the record declares none.

  • Not measured
    Standard auth

    Auth is not required, so the OAuth metadata check does not apply.

  • Not measured
    Free tier cited

    No publisher quote has been checked.

Tools

  • Pass
    Tools described

    Every tool description is at least 20 characters.

  • Pass
    Valid input schemas

    Every inputSchema is a JSON Schema object.

  • Fail
    Parameters documented

    43 of 61 parameters have a description (70%). The bar is 80%.

  • Pass
    Safety hints

    Every tool sets readOnlyHint or destructiveHint.

Publisher

  • Pass
    Verified namespace

    Listed in the Official MCP Registry.

  • Not measured
    Website matches publisher

    This namespace is not a reverse domain, so the website comparison does not apply.

  • Fail
    Source repo linked

    No repository URL is on file.

  • Fail
    Repo owner matches

    The GitHub repository owner does not match the namespace login.

  • Pass
    Recently updated

    Updated 2026-10-01T08:06:58.160481Z, within 180 days.

  • Not measured
    Open-source licence

    No GitHub repository to ask about a licence.

Start using ExtensionDash

Connect to this server

Choose your client, copy the configuration, then enable the server in your client’s MCP settings.

Run this command in your terminal, then restart Claude Code.

Claude Code

claude mcp add --transport http extensiondash https://extensiondash.com/mcp

Capabilities

Tools 16

add_competitor

Start tracking a competitor for an extension. Takes the store and id of a search_catalog row (as external_id), so a listing you just found can be promoted without looking anything up in between. Promoting a listing is what earns it a store fetch: its description, install count, rating and languages start being recorded, and its position is kept even on days it falls out of the results entirely. Where it ranks has been recorded all along, from the searches this account already runs — so a competitor added today arrives with history rather than starting from zero. The description is not available when this returns. Call list_competitors with the same extension_id until this competitor's status turns from "pending" to "scraped". At most 5 competitors per store on the Free plan and 20 on Premium, because each costs a fetch of that store on every refresh. The two stores are counted separately. On plan_limit_reached explain the cap and offer upgrading, or ask which competitor in that store the user wants removed. Never remove automatically or retry unchanged. Needs an ExtensionDash account: sign in from your app, e.g. `claude mcp login extensions-dashboard` or `codex mcp login extensions-dashboard`.

Discovered
add_extension

Track the user's own extension from its Chrome Web Store and/or Firefox Add-ons URL. Resolve a name with search_catalog first; pass the selected result's url. Do not guess between ambiguous listings or start tracking research results without the user's request. Supply both URLs in one call only when they are the same product: they share one extension_id and consume one plan slot. Repeating an existing URL is a no-op, including at the cap. If one URL matches existing tracking, its missing store is attached and existing keywords are adopted. Conflicting listings are refused, never replaced or merged. Free allows 1 tracked extension; Premium allows 10. The response includes current usage and limits. A plan_limit_reached error is not retryable unchanged: explain it, offer the pricing link, or ask which existing extension the user wants removed with remove_extension. Never delete tracking automatically to free a slot. Cached metrics may be available immediately; fresh metrics run in the background. Use get_extension and list_scrape_runs to check progress; do not promise a completion time. Needs an ExtensionDash account: sign in from your app, e.g. `claude mcp login extensions-dashboard` or `codex mcp login extensions-dashboard`.

Discovered
add_keyword

Start tracking a search term for an extension. The term is added once per store the extension is listed in, and a rank fetch is enqueued for each store that was not already tracking it. The position is not available when this returns. Call list_keywords with the same extension_id until this term's status turns from "pending" to "scanned", or list_scrape_runs to watch the fetch itself — a fetch that fails shows up there, rather than as a position that never arrives. A term is tracked per store language. Tracking "ad blocker" in en and again in de gives two independent keywords with their own positions. Call list_keywords for the languages an extension already tracks. An extension tracks at most 10 keywords on the Free plan and 50 on Premium, counted per language: "ad blocker" in en and de is two. Past that this returns a limit error. Explain the limit and offer upgrading, or ask which keyword the user wants removed with remove_keyword. Never remove automatically or retry unchanged. Needs an ExtensionDash account: sign in from your app, e.g. `claude mcp login extensions-dashboard` or `codex mcp login extensions-dashboard`.

Discovered
get_extension

One extension in full: every store listing's latest numbers plus its tracked keywords and their current positions. Each listing's rank is its position by users within its store, computed from the current user count; ties share a rank (rank_tied = others at the same count); rank_change is the move since last week's recorded rank. locales lists every store language this extension tracks, with a keyword count each, so the keywords can be worked a language at a time. Use get_metrics_history or get_keyword_ranks for movement over time. This is the tracked view, built from what we have recorded. For an extension this account does not track, use get_listing instead. Needs an ExtensionDash account: sign in from your app, e.g. `claude mcp login extensions-dashboard` or `codex mcp login extensions-dashboard`.

Discovered
get_keyword_ranks

Daily search position history for an extension's tracked keywords, one series per term per store. Every point carries the depth it was scanned to, because a position only means something against the size of the pool it came from: #1 of 9 and #1 of 50 are different results. A null position means we searched that deep and this extension was not in the results. A term tracked in more than one store language comes back as one series per language — the same term is a different SERP in each. Each series has a source. "store" is the store's own search. "google" is where the listing page ranks in Google: organic results only (ads and other blocks are not counted), top 50, desktop, in the keyword language's country (English: the United States). A null Google position means Google was searched 50 deep and the listing was not there. Needs an ExtensionDash account: sign in from your app, e.g. `claude mcp login extensions-dashboard` or `codex mcp login extensions-dashboard`.

Discovered
get_keyword_serp

The search page for one of an extension's tracked keywords, as this account last recorded it — who was on it, in what order, and which way each of them moved over the past week. Read from recorded history, not from the store: no fetch, no waiting, no egress spent. Use this when you want movement, or the same page the dashboard is showing. Every row carries mine and competitor flags, so "who is above me and which of them am I already tracking" is one call. Positions always travel with depth_scanned, because #1 of 9 and #1 of 50 are different results. Needs an ExtensionDash account: sign in from your app, e.g. `claude mcp login extensions-dashboard` or `codex mcp login extensions-dashboard`.

Discovered
get_listing

One extension's full listing from the catalog -- any extension, tracked by this account or not: description, users, rating, reviews, version, languages, categories, permissions and publisher. Pass the store and id of a search_catalog row as store and external_id. rank is the position by users within the store; ties share a rank (rank_tied = others at the same count); recomputed weekly on rank_on; rank_change is the move since the week before, positive = up. English copy, refreshed weekly from the store; refreshed_at says when. This never fetches, never answers "pending" and never starts tracking. users, rating and reviews are null when the store shows no number, which is not zero. For an extension this account tracks, get_metrics_history has its numbers over time. Shares search_catalog's 60 calls a minute per account. Works without an account too.

Discovered
get_metrics_history

Daily-or-sparser history of an extension's numbers: user count, rating and ratings count, each as a map of ISO date to value. "listings" is this extension's own store listings; "competitors" is its roster from list_competitors. store and days filter both. A map holds only dates actually observed; an empty map means no measurement in range, never zero. If users_rounded_in_range is true the store published rounded counts somewhere in the window, so small movements there may be rounding. Competitors are captured every few days rather than daily, and not on a fixed schedule, so compare a rival against "listings" on dates present in both, never by position in the map. Join a competitor entry back to list_competitors on (store, external_id); name is a label, not an identifier. include_competitors adds every roster competitor's history. competitors narrows that to specific (store, external_id) pairs and implies include_competitors; a pair not on the roster fails the whole call rather than silently dropping it. Needs an ExtensionDash account: sign in from your app, e.g. `claude mcp login extensions-dashboard` or `codex mcp login extensions-dashboard`.

Discovered
list_competitors

Who this extension competes with, measured across every keyword it tracks. Two sets. "competitors" is the roster you promoted with add_competitor: their descriptions, install counts and ratings are recorded, and their positions are kept even on days they fall out of the results. "candidates" — returned only with include_observed — is everyone else who has appeared in those searches, ranked by how often they beat you. Both are read from positions this account already recorded, not from the store: no fetch, no waiting, and each row carries the evidence behind it (how many of your terms, average position, first seen). This is the tool for deciding who is worth promoting. Needs an ExtensionDash account: sign in from your app, e.g. `claude mcp login extensions-dashboard` or `codex mcp login extensions-dashboard`.

Discovered
list_extensions

Every extension this token's owner tracks, with each store listing's latest numbers: user count (and whether the store rounded it), the 7-day change, rating, ratings count and category rank. Start here when this connection has an account: every tracking tool takes an extension_id from this list. Includes effective plan limits and extension usage (null caps mean unlimited). If empty, help the user choose a store listing and call add_extension when asked. Needs an ExtensionDash account: sign in from your app, e.g. `claude mcp login extensions-dashboard` or `codex mcp login extensions-dashboard`.

Discovered
list_keywords

Every search term tracked for one extension, one row per term per store. status is "pending" until a fetch has landed since the term was tracked, and "scanned" once one has. A "scanned" row with a null position means we searched depth_scanned results and this extension was not among them — that is a real observation, not missing data. A "pending" row can still carry a position. Removing a keyword stops the tracking but keeps the history, so a term tracked before and added again arrives with its previous result already in place; captured_on says how old that is. Poll until the status reads "scanned" to know the number answers the current tracking. locale is the store language the term was searched in. The same term ranks differently under hl=de than under hl=en, so two rows sharing a term and a store are two different results. locales lists every store language this extension tracks, with a keyword count each — start there to work a language at a time. It always reports every language, including when the locale argument narrows the rows to one. search_volume is Google Ads' average monthly searches for the term in the row's store language: for English, in the United States or worldwide as the account's profile setting says (United States unless changed), and in one country for the others (Germany for de, Spain for es, Brazil for pt-BR, and so on). It is web search demand, not searches inside the store, and is the same for both stores. null means not fetched yet, Google has no data, or the term can't be looked up (symbols such as + or %, over 80 characters or 10 words, or a store language Google Ads has no country for, such as ru); search_volume_fetched_at says which. Needs an ExtensionDash account: sign in from your app, e.g. `claude mcp login extensions-dashboard` or `codex mcp login extensions-dashboard`.

Discovered
list_scrape_runs

Recent scrape attempts against your own extensions, newest first, with their outcome and any error. Covers both the listing fetches and the rank fetches for the search terms they track. Use this after add_keyword to watch the rank fetch: the run is targeted at the search term, and a failed fetch appears here as a failure rather than as a position that never arrives. Needs an ExtensionDash account: sign in from your app, e.g. `claude mcp login extensions-dashboard` or `codex mcp login extensions-dashboard`.

Discovered
remove_competitor

Stop tracking a competitor for an extension. Stops the store fetches. The recorded history is kept, so adding the same competitor again later restores its charts rather than starting from zero — and its positions carry on being observed either way, because they come from searches this account already runs. Needs an ExtensionDash account: sign in from your app, e.g. `claude mcp login extensions-dashboard` or `codex mcp login extensions-dashboard`.

Discovered
remove_extension

Stop tracking one of this account's extensions, using an extension_id from list_extensions. Remove only the extension the user asked to remove; clarify ambiguous names first. Never remove an extension automatically to make room at a plan limit. This deletes this account's extension record, notes, keyword tracking and competitor configuration for it, across both stores. Re-adding does not restore that configuration. Shared store listings, collected metrics/rank history and other users' tracking remain. A missing or already removed id returns the same error as another account's id. Needs an ExtensionDash account: sign in from your app, e.g. `claude mcp login extensions-dashboard` or `codex mcp login extensions-dashboard`.

Discovered
remove_keyword

Stop tracking a search term for an extension, across every store it was tracked in. Positions already recorded are kept; only the tracking stops. Other extensions tracking the same term are unaffected, including ones that track the same store listing: every extension's keywords are its own. A term is tracked per store language. Tracking "ad blocker" in en and again in de gives two independent keywords with their own positions. Call list_keywords for the languages an extension already tracks. Needs an ExtensionDash account: sign in from your app, e.g. `claude mcp login extensions-dashboard` or `codex mcp login extensions-dashboard`.

Discovered
search_catalog

Search one store's catalog of extensions fetched so far, for market and niche research: store "chrome" (default) or "firefox", one per call -- Chrome users are a rounded weekly count, Firefox's an exact daily average, so the two are never mixed. listed and known in each answer show how much of that store the catalog covers. rank is the position by users within the store; ties share a rank (rank_tied = others at the same count); recomputed weekly on rank_on; rank_change is the move since the week before, positive = up. Filter by name words (match "all" or "any"; case and punctuation are ignored, and an exact name comes first unless you sort), by one category of that store (a slug such as "productivity/developer" or "privacy-security"; an unknown one is refused with the store's list) and by inclusive ranges on users, rating, reviews, language count and description length; sort by any of those or by last update. This never fetches the store. users, rating and reviews are null when the store shows no number, which is not zero. Each row says whether it is one of yours or a competitor you track; pass its store and id to add_competitor to start tracking it. For one extension's full listing, call get_listing. Allows 60 calls a minute per account. Works without an account too.

Discovered

Recent measurements

Probe history

View status
Latest stored probes. Raw probe rows are kept for 14 days.
CheckedResultResponseTools
Connected603 ms16