atlassian-marketplace-mcp (rustem-shiriiazdanov/atlassian-marketplace-mcp) is an MCP server listed on the M8ven Trust Index. It scores 74 out of 100, grade C. It declares 91 tools. No publisher has claimed this listing.

C
Caution
74/100

atlassian-marketplace-mcp

Exposes Atlassian Marketplace vendor APIs as 95 MCP tools, 3 resources, and 4 prompts, enabling LLMs to manage licenses, promotions, reviews, and reporting metrics without writing API calls.

Caution. Specific findings reduced this grade. They are listed on the page. Grades reflect the full trust pyramid: code, verification depth, and reputation. New projects cap at C until adoption is earned.

How we verified

Code Verified⚡ Live Monitored: not connected

Verified is a snapshot. Live keeps it current, and builds your track record.

⚡ Connect GitHub → continuous verification on every pushwhy connect →

Who stands behind it

rustem-shiriiazdanov

Source: Glama

Is this your MCP?

Claim it to get a verified publisher badge, a free copy of our full audit findings, and direct contact for any high-priority issues we find. Or connect your repo for our deepest verification, Live Monitored: read-only, revoke anytime. What we access →

Install from

The grade above is for the source repository. Registries can serve a different version, so we mark the ones we were not able to read.

// key findings
⚠️
Known vulnerabilities in dependencies: 1 high
Affects packages this MCP installs at runtime. Upgrade or remove the affected dependency.
⚠️
Tests do not pass
Either the test suite is broken or the code regressed. Either way the published behaviour can’t be verified by the publisher’s own tests.
No credential exfiltration, no sensitive file access, no obfuscation
Static analysis found nothing flowing your secrets to unexpected places.
Open source with a license and README
Anyone can audit the code, the license is declared, and the publisher documents what it does.
// tools this server exposes91 tools

These names and descriptions are the publisher's own, read from the source code. We print them as written. Our assessment is the findings above, not this list.

app_listing_get

Get Marketplace product listing metadata for one app. Returns `{productId, appKey, developerId, appName, summary, tagLine, images, tags, communityEnabled, developerLinks, thirdPartyIntegrations, state, approvalStatus, approvalDetails, slug, cloudComplianceBoundary, hostingVisibility, marketingLabels

app_listing_update

Update Marketplace product listing metadata. PUBLIC IMPACT: changes appear on the app's marketplace page after approval. PUT semantics — body should be a full listing object.

app_requests_and_approvals

Marketplace 'request app' / 'approve app' activity, per month. Benchmark-style aggregate (NOT a paginated list): `{total:{name, appRequestsAndApprovalsPerMonth:[{date, appRequests, appRequestsApproved, appRequestsApprovalRate}]}, addons:[{addonKey, name, productId, appRequestsAndApprovalsPerAppPerMo

app_software_get_by_appkey

Look up app-software (the technical artifact behind a product listing) by its appKey. Returns an ARRAY of `{appSoftwareId, hosting, complianceBoundaries, archived}` — one entry per hosting platform the app supports. `complianceBoundaries` is a Cloud-only concept (an array like `["commercial"]` for c

app_software_versions_list

List versions for an app-software. Returns `{links, versions:[{buildNumber, versionNumber, compatibilities, supportedPaymentModel, frameworkDetails, licenseType, ...}], totalCount}`. CURSOR-paginated (`limit`+`cursor` from `links.next`) — `offset` is NOT supported (silently ignored).

app_software_version_create

Create a new version for an app-software. PUBLIC IMPACT (eventually): once a version is approved and listing is published, customers can install it.

app_software_version_get

Get one version of an app-software by build number.

app_software_version_update

Update one version of an app-software (PUT — full replace).

app_software_tokens_list

List API access tokens minted for this app-software. Returns `{tokens:[{token, cloudId, instance}]}` — each token maps to one Cloud install. CREDENTIAL-ADJACENT: exposes token identifiers + the customer cloud sites they belong to.

app_software_token_create

Create a new access token for this app-software. CREDENTIAL: the returned token must be stored securely — Atlassian will not show it again.

app_version_listings_list_all

List version-listings for an app-software (per-version published metadata: screenshots, highlights, moreDetails, youtubeId, developerLinks, approvalStatus, state, buildNumber, revision). Returns `{links, versions:[…]}`. Despite the name, it's CURSOR-paginated (default 10/page; pass `cursor` from `li

app_version_listing_get

Get the version-listing for a specific build number.

app_version_listing_create

Create a new version-listing for a specific build. PUBLIC IMPACT (after approval): publishes a new app version to the Marketplace.

app_version_listing_update

Update an existing version-listing (PUT — full replace). PUBLIC IMPACT: changes the customer-facing version metadata after approval.

apps_list

Discover apps in this developer space (live API call). Returns productId (UUID — use as filter), appKey, appName, state. CURSOR-paginated: pass `cursor` (from a prior response's `nextCursor`) to page forward. In summary mode the result includes `nextCursor` (null when no more pages). Default (no `li

apps_known

Return the static name -> productId map loaded from PRODUCT_ID_* env vars. Use this to look up product UUIDs by friendly name without an API call.

artifact_fetch_from_url

Have Atlassian fetch an artifact from a public URL and store it. Returns `{fileInfo, _links, details}` — the stored artifact's id/download link live under `_links`/`fileInfo`. NOTE: the API field is `uri` (not `url`); this tool accepts `url` and maps it to `uri` for you.

artifact_get

Get artifact metadata (name, size, content type, download URL).

benchmark_sales

Sales benchmark vs. ecosystem, per month. Returns `{total:{name, salesBenchmarkPerMonth:[…]}, addons:[{addonKey, name, productId, salesBenchmarkPerMonth}]}`. Each month row: `{date, sale, previousMonthSale, salesMoMGrowth, salesPercentile, salesMoMGrowthBenchmarkAllPartners, salesYTD, salesYTDLastYe

benchmark_evaluations

Evaluations benchmark vs. ecosystem, per month. Returns `{totals:{name, evaluationBenchmarkPerVendorPerMonth:[…]}, addons:[{addonKey, name, productId, evaluationBenchmarkPerAppPerMonth}]}`. NOTE the wrapper is `totals` (plural) and the per-month key differs between total level (`…PerVendorPerMonth`)

developer_space_by_vendor

Resolve a developerId from a vendorId (legacy mapping).

developer_space_get

Get developer space profile by developerId. Returns `{id, vendorId, name, status, type, organisationId, version}`. Defaults to MARKETPLACE_DEVELOPER_ID. Unknown/malformed id → HTTP 400.

developer_space_catalog_account

Get the catalog-account ID for a developer space (used by some downstream services).

developer_space_listings

Get the developer-profile listing documents (developer profile copy, web metadata, etc.). Returns a bare ARRAY of listing objects (not an envelope). NOT the product apps — use apps_list for those.

developer_space_members_list

List team members in the developer space. Returns `{members:[{aaid, roles, categories, email, userName}], next}`. CURSOR-paginated, but with a NON-STANDARD shape (unlike the `links.next` URL used by other list tools): `next` is a BARE opaque cursor token (or absent on the last page) — pass its value

developer_space_member_get

Get one developer-space team member by Atlassian account id (aaid). Returns `{aaid, roles, categories, email, userName}` — contains PII (email, userName). Unknown aaid → HTTP 400.

developer_space_member_add

Add a user to the developer space. AFFECTS OTHERS: grants console access to this user.

developer_space_member_update

Update a developer-space team member (PUT — full replace).

developer_space_member_remove

Remove a user from the developer space. AFFECTS OTHERS: revokes their console access.

evaluations_by_metric

Evaluation time-series grouped by a dimension. FLAT `total.series[]` (no datasets, no uniqueTotal) — one series per group value (e.g. country names for `metric=country`), each `{name, elements:[{date,count}]}` — plus per-app `addons[]`. Only aggregation/startDate/endDate filter (productId/hosting/ad

feedback_details

Raw customer feedback entries (uninstall/disable/unsubscribe events with reasons + free-text messages). Returns `{_links:{self,query,next}, feedback:[{addonKey, addonVersion, applicationKey, applicationVersion, hosting, date, feedbackType, reasonKey, reason, message, fullName, appEntitlementId, appE

feedback_metrics_by_metric

Feedback time-series grouped by a metric. FLAT `total.series[]` (no datasets, no uniqueTotal) — one series per group value, each `{name, elements:[{date,count}]}` — plus per-app `addons[]`. For `metric=reason` series are reasonKeys (bugs, merging, not-meeting-needs, other, project-based, sandbox, us

free_starter_tier_export

Export Cloud free-starter-tier entitlements (free users on your apps) as of a single date. Returns JSON array of `{day, licenseId, appEntitlementId, entitlementNumber, parentEdition, dateOfEvaluation, parentUnitCount, technicalEmail, vendorId, addonName, addonKey, productId}` by default; pass `accep

licenses_list

List licenses for the developer space (paginated, offset/limit). Use 'text' to find a single license by SEN.

licenses_export_sync

Synchronous export of licenses. `accept=csv` (default) returns a CSV string; `accept=json` returns a JSON array. May 5xx on large ranges — prefer the async variant. Request timeout is bumped to 10 min (overridable via EXPORT_TIMEOUT_MS env).

licenses_export_async_start

Start an async license export job. Returns `{export:{id}}` to poll. `accept=csv|json` sets the format the eventual download will produce (the start response itself is always the id envelope).

licenses_export_async_status

Poll status of an async license export job.

licenses_export_async_download

Download a completed async license export. Returns the JSON array of license records. Request timeout is 10 min (overridable via EXPORT_TIMEOUT_MS env) so multi-MB downloads don't get aborted mid-stream. Only call this after the *_status endpoint reports DONE — otherwise you'll get a 404 'Export is

marketing_attribution_export_async_start

Start an async export of marketing-attribution data. Returns `{export:{id}}` — pass that id to `marketing_attribution_export_async_status` then `_download`. Filters: productId/addon/text/startDate/endDate (+ `accept` for the eventual download format). Export is param-deduped: identical params yield

marketing_attribution_export_async_status

Poll status of an async marketing-attribution export job. NOTE the generic /async/export/ path (shared, not prefixed).

marketing_attribution_export_async_download

Download a completed async marketing-attribution export. Returns JSON records. Request timeout is 10 min (overridable via EXPORT_TIMEOUT_MS env). Only call this after the *_status endpoint reports DONE — otherwise you'll get a 404 'Export is being processed'.

cloud_migration_compat_get

Get DC-to-Cloud migration compatibility info for an app. Returns `{developerId, productId, addonKey, addonName, cloudMigrationAssistantCompatibility, migrationPath, isDualLicenseOptedIn}`. NOTE: returns HTTP 404 (surfaced as an error) for apps that have no migration-compatibility record configured —

cloud_migration_compat_create

Create cloud-migration compatibility info for an app. PUT semantics — body is the full document.

cloud_migration_compat_update

Patch cloud-migration compatibility info (partial update).

parent_software_list

List parent software (the Atlassian products your apps target — Jira, Confluence, Bitbucket, etc.). Returns `{links, parentSoftware:[…]}`. CURSOR-paginated (`limit`+`cursor` from `links.next`); the default page returns all ~23 products.

parent_software_get

Get one parent software (Atlassian product) by ID (e.g. `jira`, `confluence`). Returns `{id, developerId:'Atlassian', name, hostingOptions:[{hosting}], extensibilityFrameworks, state, revision}`. Nonexistent id → HTTP 404.

parent_software_versions_list

List known versions of a parent software (e.g. Jira versions Atlassian has published). CURSOR-paginated (`limit`+`cursor` from `links.next`).

parent_software_version_by_build

Get a parent-software version by its build number (path `/versions/build/{buildNumber}`). Returns `{buildNumber, versionNumber, hosting:[…], state, revision, createdAt}`. Unknown build → HTTP 404.

parent_software_version_by_number

Get a parent-software version by its human-readable version number (path `/versions/number/{versionNumber}`, e.g. '11.3.8'). Same record shape as version_by_build (`{buildNumber, versionNumber, hosting, state, revision, createdAt}`). Unknown version → HTTP 404.

partner_metrics_fetch

Fetch partner-metric time series (POST). The `body` shape is `ReportingMetricTimeSeriesRequestBody`: `{metrics:{metricSets:[…], metricFields:[…]}, dateRange:{startDate, endDate}, granularity:'YEAR'|'MONTH'|'WEEK'|'DAY', attributes?, sortByList?, attributeFilter?}`. IMPORTANT: `metrics` is an OBJECT

privacy_security_get

Get privacy & security information for an app (used by enterprise procurement reviewers). Returns `{commonCloud:{dataAccessAndStorage, logDetails, dataResidency, privacy, security, properties, hasRestAPIExtension, supportsConfigurableEgress}}`. `state=live` (default) returns the published version; `

privacy_security_draft_put

Create or update the draft privacy-and-security information (not yet public). PUT — full replace.

privacy_security_draft_delete

Delete the draft privacy-and-security info (the currently published version is unaffected).

privacy_security_publish

Publish the current draft privacy-and-security info. PUBLIC IMPACT: this version becomes visible to all Marketplace visitors and procurement reviewers.

product_catalog_latest

Get a presigned S3 URL for the latest public Marketplace app-catalog snapshot. Response shape: {date, presignedUrl, expiresInSeconds}. The presignedUrl points to a LARGE CSV file (~150 MB, Content-Type binary/octet-stream — NOT JSON), one row per published app with columns like: is_beta, summary, ta

promotions_list_paged

List promotions (paginated). STRONGLY prefer this over promotions_list (the non-paged variant can time out). Returns `{_links, promotions:[…], offset, limit, totalItems, orderBy, nextId, prevId}`. Cloud uses cursor pagination via `nextId`/`prevId` (and `totalItems` is null); Server/DC uses `offset`/

promotions_list

List ALL promotions in one non-paginated response (legacy). WARNING: on partners with many promotions this endpoint is very slow and can hit the request timeout (60s, then retried) — effectively hanging. Prefer promotions_list_paged in almost all cases; use this only when you truly need every promot

promotions_create

Create a new promotion. PUBLIC IMPACT: promo code becomes redeemable by customers. Required: name, eligibleApps, expirationDate, hostingType, promotionType, discountType.

promotions_get

Get one promotion by ID. Returns the full promotion object (~21 fields: id, name, eligibleApps, startDate, expirationDate, status, promotionType, discountType, discountPercent, maxUses, used, hostingType, promotionCode, …). GOTCHA: a nonexistent/malformed promotionId returns HTTP 500 (not 404).

promotions_update

Update a promotion (PATCH — only supplied fields change).

promotions_status

Get the status of a promotion (ACTIVE | ENDED_EARLY | EXPIRED).

promotions_codes_list

List single-use codes for a SINGLE_USE_PROMOTION.

promotions_codes_create

Generate a new single-use code for a promotion.

promotions_code_get

Get one single-use code (including usage info if redeemed).

promotions_code_delete

Delete an unused single-use code.

reporting_links

Get the reporting root — a HAL response that lists all available reporting links for this developer space.

reviews_list

List customer reviews for an app. CURSOR-paginated (not offset). Returns `{productId, reviews:[{id, content, stars, date, totalVotes, helpfulVotes, productHosting, isFlagged, authorName, transitionedToFiveStarRating}], cursor, count, averageStars}` where `count` is the total review count, `averageSt

review_get

Get a single review by ID.

review_response_put

Post or update a vendor response to a review. PUBLIC IMPACT: response is visible to all Marketplace visitors.

review_response_delete

Delete the vendor's response to a review. PUBLIC IMPACT: removes a publicly visible response.

metrics_churn

Cloud churn TIME-SERIES (not a single rate). Returns `total.datasets` split by billing period (`Monthly`, `Annual`) with two series each: `Customers` (cohort denominator) and `Cancellations` (numerator), plus per-app breakdown in `addons[]`. Caller computes rate = Cancellations / Customers per bucke

metrics_churn_benchmark

Per-app monthly churn benchmark vs. ecosystem average. Returns `churnBenchmarkPerApp[]` where each entry has `churnBenchmarkPerMonth[]` rows: `{year, month, churnedLicenses, totalLicenses, churnRate, isolatedChurnRate, churnRateBenchmark, isolatedChurnRateBenchmark}`. The `*Benchmark` fields are nor

metrics_conversion

Cloud evaluation→paid conversion TIME-SERIES. Shape differs from churn/renewal: `total.series[]` is FLAT (no `datasets[]` billing-period split) with two series — `Evaluations` (denominator) and `Conversions` (numerator) — each a list of `{date, count}` elements. No `uniqueTotal` field. `addons[]` ca

metrics_renewal

Cloud renewal TIME-SERIES. `total.datasets[]` split by billing period (`Annual`, `Monthly`) — like churn — with two series each: `Renewal opportunities` (denominator) and `Renewals` (numerator). NOTE: unlike churn, renewal series have NO `uniqueTotal` field (each series is just `{name, elements:[{da

metrics_details_by_metric

License-event details underlying a sale metric. Returns `events[]` rows: `{addonKey, addonName, hosting, lastUpdated, eventDate, transactionId, licenseDetails, productId}`. Supports rich filters (addon, hosting, partnerType, text, sortBy, order, offset, limit). Server caps limit at 50.

metrics_details_export

Export of license-event details for a sale metric. `accept=csv` (default) returns the 17-column CSV (`addonName,addonKey,hosting,lastUpdated,eventDate,transactionId,licenseId,maintenanceStartDate,maintenanceEndDate,monthsValid,appEntitlementId,appEntitlementNumber,cloudId,inGracePeriod,multiInstance

search_keywords_partner

Top search keywords across all the partner's apps (developer-space wide). Returns `{total:{searchAppearances, topSearchKeyword}, addons:[{addonName, addonKey, productId, leadingSearchKeyword, searchAppearances, elements:[{date,count}]}]}`. Filters: aggregation/startDate/endDate only (pagination/prod

search_keywords_partner_export

Export variant of partner-wide search keywords. With `accept=json` (default) returns the data INLINE as `{_links:{self,query,export}, total, addons}` (same payload as `search_keywords_partner`); with `accept=csv` returns a per-app CSV string. ⚠️ The advertised `_links.export` download URLs are BROKE

search_keywords_by_source

Top search keywords for one source. Returns `{details:[{searchKeyword, percentage}]}` (flat keyword share, no time series). Filters: startDate/endDate (no aggregation; pagination ignored).

search_keywords_by_source_export

Export variant of source-filtered search keywords. UNLIKE the partner/by_app exports, this returns the FULL DATA directly as a JSON array of `{searchKeyword, percentage}` rows (up to 500), NOT HAL download links. Large responses spill to the truncation envelope.

search_keywords_by_app

Top search keywords for one app. Returns `{summary:{addonName, addonKey, leadingSearchKeyword, …}, details:[{searchKeyword, keywordCount, elements:[{date,count}]}]}`. `productId` is a PATH segment. Filters: aggregation/startDate/endDate (pagination/hosting ignored).

search_keywords_by_app_export

Export variant of per-app search keywords. Returns the data INLINE as `{_links:{self,query,export}, summary, details}` — same payload as `search_keywords_by_app`. ⚠️ The advertised `_links.export` CSV/JSON download URLs are BROKEN (Atlassian-side doubled `/export/export` path → 404, verified 2026-06

zero_search_results_keywords

Keywords that produced ZERO search results, for one source — SEO gap analysis. Returns `{details:[{searchKeyword, count}]}` (count = how many times the no-result search happened). Filters: startDate/endDate (no aggregation; pagination ignored).

zero_search_results_keywords_export

Export variant of zero-result keywords. Returns the FULL DATA directly as a JSON array of `{searchKeyword, count}` rows (up to 500), NOT HAL download links (same as the by_source export). Large responses spill to the truncation envelope.

transactions_list

List sales transactions for this vendor's apps (refunds appear inline as negative amounts). NOTE per Atlassian: this endpoint can return 5xx on large datasets — for full pulls, prefer transactions_export_async_start + status + download. Use 'text' to find by transactionId, licenseId, SEN, customer i

transactions_export_sync

Synchronous export of transactions. `accept=csv` (default) returns a CSV string; `accept=json` returns a JSON array. Prefer async for large ranges. Request timeout is 10 min (overridable via EXPORT_TIMEOUT_MS env).

transactions_export_async_start

Start an async transactions export job. Returns `{export:{id}}` to poll. `accept=csv|json` sets the eventual download format (start response is always the id envelope).

transactions_export_async_status

Poll status of an async transactions export job.

transactions_export_async_download

Download a completed async transactions export. Returns the JSON array of transaction records. Request timeout is 10 min (overridable via EXPORT_TIMEOUT_MS env). Only call this after the *_status endpoint reports DONE — otherwise you'll get a 404 'Export is being processed'.

transactions_aggregate_by_metric

Aggregated sales grouped by a metric path segment. Maps to /sales/transactions/{metric}.

transactions_aggregate_by_hosting

Friendly alias for transactions_aggregate_by_metric(metric='hosting'). HAL template `hosting{?aggregation,startDate,endDate}`. NOTE: `productId` is silently ignored on this specific endpoint (verified 2026-06-03) — use `transactions_aggregate_by_metric` with `metric=hosting` if you need productId sc

// known CVEs in dependencies1 high

Disclosed vulnerabilities in this server's declared npm dependencies (via OSV). Whether each is reachable depends on the installed versions.

high@modelcontextprotocol/sdk@1.0.0GHSA-w48q-cv73-mx4w

Model Context Protocol (MCP) TypeScript SDK does not enable DNS rebinding protection by default

Depend on this server? Get alerted when its CVEs change.Watch this server free →
// environment variables
To run this server yourself, you supply these values. They go in your own MCP client configuration and stay on your machine. The secret label means the value is sensitive, not that the server mishandles it.
configEXPORT_TIMEOUT_MS
configHTTP_MAX_ATTEMPTS
configHTTP_TIMEOUT_MS
configMARKETPLACE_PARTNER_IDNumeric partner ID for the v1 promotions API
configMAX_RESPONSE_CHARS
configSWAGGER_PATH
// quality suggestions

Tool annotations

No tools have read-only/destructive annotations

Add readOnlyHint or destructiveHint annotations to every tool so hosts can warn users before invoking.

All four hints declared on every tool

91/91 tools missing one or more hints — app_listing_get (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); app_listing_update (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); app_requests_and_approvals (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint), +88 more. OpenAI's directory rejects tools where any of the four hints are missing or non-boolean.

For every tool, set all four hints (readOnlyHint, destructiveHint, idempotentHint, openWorldHint) to explicit true/false values that match the handler’s actual behaviour.

Tool inputs are validated

65/91 tool handlers declare input schemas (71%)

Declare an inputSchema with zod/joi/yup on every tool definition.

Tool handlers catch errors

Only 0/91 tool handlers wrap calls in try/catch (0%)

Wrap each tool handler body in try/catch and return a structured error response.

Tool test coverage

63/91 tools referenced in tests (69%)

Write tests that reference each tool by name so every tool has at least one test.

Tests pass

npm test failed — tests do not pass

Make sure npm test runs cleanly. Common cause: missing build step or missing env vars.

Shell command execution

2 child_process calls — runs shell commands

Prefer library functions over shell-outs. If you must shell out, ensure all inputs are properly escaped.

Production dependencies are patched

0 critical, 1 high severity in production deps — @modelcontextprotocol/sdk@1.0.0 (high)

Run npm audit fix, or upgrade the affected packages to a non-vulnerable version.

Claim the listing to review these findings one by one and send us a correction where you disagree, straight to the team. Claiming also means we tell you when the grade moves, and reach you first if we find anything urgent.

// full audit trail
The findings above are the summary. The full trail, every check we ran, each deduction, the network hosts observed and the dependency advisories, goes to verified publishers, along with an alert whenever a new one lands. Verified publishers can also review each finding and dispute it in one click. Publisher corrections have sharpened several of our checks this month, because the maintainer knows the codebase better than any scanner.
// improvement guidance — verified publishers only
We have 8 concrete improvements we can share with the publisher of this MCP. Each comes with specific guidance to raise the trust score.
// embed badge in your README
[![M8ven Verified](https://m8ven.ai/badge/mcp/rustem-shiriiazdanov/atlassian-marketplace-mcp?variant=verified)](https://m8ven.ai/mcp/rustem-shiriiazdanov/atlassian-marketplace-mcp)
Shows verification status without the grade. Want the grade badge instead? Remove ?variant=verified from the URL.
commit: ecdf9d3d8323273a44d10024962322591c51cbda
code hash: 3c50348c2717d7c3cab6690dbd9a3e4612237355ea591429ce1f1266bf908984
view raw JSON →
Check MCPs from inside your assistant
Tool Check · MCP

Vetting this one by hand? Tool Check is an MCP that scores other MCPs. Add it once and ask Claude, ChatGPT, or any MCP client to grade a server, surface CVEs, check the publisher, and suggest safer alternatives — before you install.

https://m8ven.ai/api/mcp/tool-check
check_toolsearch_toolscompare_toolsrecommend_alternativescheck_publisherreport_concern
How to add it →Free · no account needed · works in any MCP client