Pixeloa MCP server
ai.pixeloa/pixeloapixeloa.aistreamable-httpv1.1.0Scanned
Turn product photos into true-scale 3D models and AR pages, then publish, embed and catalog them.
20 tools: 4 can delete or cancel things, 1 do something that can't be undone, 6 change things and 9 only read.
- Reads private data · 0
- Reads text others write · 1
- Can send data out · 1
- No findings
Tools listed without sign-in. Remote posture
With Anlyon
4 tools here can delete or cancel things and 1 does something that can't be undone. Govern them.
- Anlyon lists the tools and adopts none.
- Each tool you adopt gets approvals and receipts.
- Your agent never holds the key.
A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.
What the Pixeloa MCP server can do
The Pixeloa MCP server is published by pixeloa.ai, a domain the MCP Registry verified. It has 20 tools. It can delete or cancel things (4 tools: deleteItem, unpublishItem, deleteCatalog and removeCatalogItem), do things that cannot be undone, such as sending, paying or publishing (publishItem) and create or change data (6 tools: createItem, updateItem, confirmScale, startGeneration and 2 more). 9 tools only read.
1 tool brings in text other people can write (getItemLinks); 1 tool can send data out of your system (publishItem).
None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.
Its remote endpoint lists its tools without sign-in.
Last scanned Nov 11, 2026, version 1.1.0.
Tools
Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.
20 tools
Deletes or cancels 4Claude Code: deny
deleteCatalogDelete a catalog (unpublishes its public page first).No findings
Why it is in this group
- The name "deleteCatalog" says "delete": a word for removing or ending something.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Delete a catalog (unpublishes its public page first). Permanent. A published catalog is unpublished first so its public page, QR and embeds go quiet; item memberships are removed (the items themselves are untouched); if it was the primary catalog that pointer clears. The platform demo catalog cannot be deleted (409 conflict). CHANGES PUBLIC STATE. Requires scope: write_catalog.
Annotations the server declares
- title: Delete a catalog (unpublishes its public page first)
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: false
deleteItemDelete an item (unpublishes first; removes images and models).No findings
Why it is in this group
- The name "deleteItem" says "delete": a word for removing or ending something.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Delete an item (unpublishes first; removes images and models). Permanent. A published item is unpublished first so its AR page and embeds go quiet, then its images, models and QR are removed. Requires the same scope as creating items. CHANGES PUBLIC STATE. Requires scope: upload_images.
Annotations the server declares
- title: Delete an item (unpublishes first; removes images and models)
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: false
removeCatalogItemRemove an item from a catalog.No findings
Why it is in this group
- The name "removeCatalogItem" says "remove": a word for removing or ending something.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Remove an item from a catalog. Requires scope: write_catalog.
Annotations the server declares
- title: Remove an item from a catalog
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: false
unpublishItemUnpublish the item's public AR page (all embeds go quiet).No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Unpublish the item's public AR page (all embeds go quiet). CHANGES PUBLIC STATE. Requires scope: publish_items.
Annotations the server declares
- title: Unpublish the item's public AR page (all embeds go quiet)
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: false
- openWorldHint: false
Can't be undone 1Claude Code: ask
publishItemPublish the item's public AR page.Can send data outNo findings
Why it is in this group
- The name "publishItem" says "publish": a word for a change that another call does not undo.
What it can reach
- outboundThe name "publishItem" says "publish": it sends or publishes something to other people.
Description, as the server wrote it Publish the item's public AR page. Makes the item public (this is the owner's publish decision; items created via the API start private) subject to the plan's published-item limit (403 plan_limit_reached), verifies eligibility (409 conflict with details.state / details.blockers when the model, scale or category policy blocks it), mints the public slug on first publish and returns the item with public.ar_url and public.embed_url set. Check item.publishable before calling. CHANGES PUBLIC STATE. Requires scope: publish_items.
Annotations the server declares
- title: Publish the item's public AR page
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
Changes things 6Claude Code: ask
addCatalogItemAdd an item to a catalog (appends unless position is given).No findings
Why it is in this group
- The name "addCatalogItem" says "add": a word for a change.
Description, as the server wrote it Add an item to a catalog (appends unless position is given). Both the catalog and the item must belong to the key's account (404 otherwise). 409 conflict when the item is not ready (details.status says where it is), is not eligible for catalogs, is unlisted (link only), or is already in the catalog. Requires scope: write_catalog.
Annotations the server declares
- title: Add an item to a catalog (appends unless position is given)
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
confirmScaleConfirm the item's real-world dimensions (cm).No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
Description, as the server wrote it Confirm the item's real-world dimensions (cm). Allowed once analysis has finished (status 'awaiting_confirmation', or 'failed' with a completed analysis) and before generation. Moves the item to 'confirmed'. Requires scope: confirm_scale.
Annotations the server declares
- title: Confirm the item's real-world dimensions (cm)
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
createItemCreate an item from 1–4 image URLs and start analysis.No findings
Why it is in this group
- The name "createItem" says "create": a word for a change.
Description, as the server wrote it Create an item from 1–4 image URLs and start analysis. Images are fetched server-side (https only, size- and type-checked against the active provider's limits). The first image is the primary/front view. Image URLs must be publicly reachable and must not point at pixeloa.ai itself (host them on your own site, a CDN, or a public bucket). The item enters 'analyzing'; poll GET /v1/items/{id} until next_allowed_actions contains confirm_scale. Requires scope: upload_images.
Annotations the server declares
- title: Create an item from 1–4 image URLs and start analysis
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
generateItemQrGenerate (or regenerate) the item's QR code SVG.No findings
Why it is in this group
- The name "generateItemQr" says "generate": a word for a change.
Description, as the server wrote it Generate (or regenerate) the item's QR code SVG. Requires scope: create_qr.
Annotations the server declares
- title: Generate (or regenerate) the item's QR code SVG
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
startGenerationStart 3D generation (spends the account's credits).No findings
Why it is in this group
- The name "startGeneration" says "start": a word for a change.
Description, as the server wrote it Start 3D generation (spends the account's credits). Requires status 'confirmed' with confirmed dimensions (a 'failed' generation may be attempted again on the same item; it was refunded and has no result). A 'ready' item is never regenerated in place; for another result create a new item from the images. Asynchronous: poll GET /v1/items/{id} until model.available is true. Fails with 402 insufficient_credits when the balance is too low, 409 conflict when the state does not allow it. SPENDS CREDITS: confirm with the account owner unless they delegated this. Requires scope: start_generation.
Annotations the server declares
- title: Start 3D generation (spends the account's credits)
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
updateItemUpdate the item's public-page settings (product CTA link).No findings
Why it is in this group
- The name "updateItem" says "update": a word for a change.
Description, as the server wrote it Update the item's public-page settings (product CTA link). Sets the 'view product' button shown on the public AR page and embeds. product_url must be a public https URL on your own site, and it may not point at pixeloa.ai itself or a private network (422 validation_error otherwise). Pass null to clear it (the label clears with it). product_cta_label (max 60 chars) overrides the default button text. Clicks are counted in the owner's analytics. Requires scope: publish_items.
Annotations the server declares
- title: Update the item's public-page settings (product CTA link)
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
Only reads 9Show 9 read-only tools
getItemGet one item, including status and next_allowed_actions.No findings
Why it is in this group
- The name "getItem" says "get": a word for reading.
Description, as the server wrote it Get one item, including status and next_allowed_actions. Requires scope: read_catalog.
Annotations the server declares
- title: Get one item, including status and next_allowed_actions
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
getItemLinksPublic AR URL, embed URL, embed snippets and QR (if generated).Reads text others writeNo findings
Why it is in this group
- The name "getItemLinks" says "get": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "url": it reads web pages, which anyone can write.
Description, as the server wrote it Public AR URL, embed URL, embed snippets and QR (if generated). Requires scope: read_catalog.
Annotations the server declares
- title: Public AR URL, embed URL, embed snippets and QR (if generated)
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
getMeWho am I, what plan am I on, and what credits do I have.No findings
Why it is in this group
- The name "getMe" says "get": a word for reading.
Description, as the server wrote it Who am I, what plan am I on, and what credits do I have. Requires scope: read_usage.
Annotations the server declares
- title: Who am I, what plan am I on, and what credits do I have
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
listAgentsList the account's agent identities.No findings
Why it is in this group
- The name "listAgents" says "list": a word for reading.
Description, as the server wrote it List the account's agent identities. An agent is an identity OWNED by the account (kind agent or service). Keys attach to an agent; the agent gets actor identity, an optional daily credit cap and its own audit trail, but every call still runs as the owner. Agents are created and suspended by a human in the dashboard. Requires scope: read_usage.
Annotations the server declares
- title: List the account's agent identities
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
listCatalogsList the account's catalogs.No findings
Why it is in this group
- The name "listCatalogs" says "list": a word for reading.
Description, as the server wrote it List the account's catalogs. Requires scope: read_catalog.
Annotations the server declares
- title: List the account's catalogs
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
listCreditLedgerThe account's credit ledger (newest first, cursor-paginated).No findings
Why it is in this group
- The name "listCreditLedger" says "list": a word for reading.
Description, as the server wrote it The account's credit ledger (newest first, cursor-paginated). Every debit and credit: generation charges, refunds, plan allocations, top-ups. Query: limit (1–100), cursor (created_at from next_cursor). Requires scope: read_usage.
Annotations the server declares
- title: The account's credit ledger (newest first, cursor-paginated)
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
listGenerationOptionsGeneration options available to this account and their current credit cost.No findings
Why it is in this group
- The name "listGenerationOptions" says "list": a word for reading.
Description, as the server wrote it Generation options available to this account and their current credit cost. Read before spending. Each option lists credit_cost (what POST /v1/items/{id}/generate will debit for that provider), release_stage, image limits and output formats; is_default marks the provider used when the item has none pinned. GET /v1/items/{id} additionally returns generation.estimated_credit_cost for that item. Requires scope: read_usage.
Annotations the server declares
- title: Generation options available to this account and their current credit cost
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
listItemsList the account's items (newest first, cursor-paginated).No findings
Why it is in this group
- The name "listItems" says "list": a word for reading.
Description, as the server wrote it List the account's items (newest first, cursor-paginated). Query: limit (1–100, default 25), cursor (created_at ISO string from next_cursor), status. Requires scope: read_catalog.
Annotations the server declares
- title: List the account's items (newest first, cursor-paginated)
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
listWebhookEndpointsList the account's webhook endpoints and the event types.No findings
Why it is in this group
- The name "listWebhookEndpoints" says "list": a word for reading.
Description, as the server wrote it List the account's webhook endpoints and the event types. Endpoints are created by a human in the dashboard (Account → API keys → Webhooks); a key can list them but never add or redirect one. Deliveries are signed with Pixeloa-Signature (t=<unix>,v1=<hex HMAC-SHA256(secret, t + "." + body)>), retried with backoff for up to 12h, and auto-disabled after 20 consecutive failures. Requires scope: read_usage.
Annotations the server declares
- title: List the account's webhook endpoints and the event types
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.
Exfil surface
Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.
No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data.
- Reads private data0
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write1
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out1
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.
Remote posture
What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.
Looked at Nov 20, 2026, 18:27 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.
- Reachable
- yes
- HTTP status of an anonymous initialize
- 200
- tools/list served without credentials
- yes
- Tools that change things, listed without credentials
- 11
- Protocol version
- 2025-06-18
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- not observed
- Authorization servers
- not observed
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- not observed
- Client ID metadata documents
- not observed
- HighThe whole server
The endpoint served tools/list without credentials, and the list has 11 tools that change things (4 destructive, 1 irreversible, 6 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://pixeloa.ai/mcp: tools/list returned 20 tools to an anonymous client, including deleteItem (destructive), unpublishItem (destructive), deleteCatalog (destructive), removeCatalogItem (destructive), publishItem (irreversible) and 6 more
Findings
Matches of fixed checks, with the text they matched.
No check matched anything in these tool definitions. That is not a safety guarantee.
History
Changes to its tool definitions and to its remote endpoint, newest first.
No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.
Install
Pinned to version 1.1.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http pixeloa https://pixeloa.ai/mcp --header 'Authorization: <your-authorization>'claude mcp add --transport http pixeloa https://pixeloa.ai/mcp --header 'Authorization: <your-authorization>'Run in a terminal.
Claude Code permissions
Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.
{
"permissions": {
"allow": [
"mcp__pixeloa__getMe",
"mcp__pixeloa__listItems",
"mcp__pixeloa__getItem",
"mcp__pixeloa__getItemLinks",
"mcp__pixeloa__listCatalogs",
"mcp__pixeloa__listGenerationOptions",
"mcp__pixeloa__listCreditLedger",
"mcp__pixeloa__listAgents",
"mcp__pixeloa__listWebhookEndpoints"
],
"ask": [
"mcp__pixeloa__createItem",
"mcp__pixeloa__updateItem",
"mcp__pixeloa__confirmScale",
"mcp__pixeloa__startGeneration",
"mcp__pixeloa__publishItem",
"mcp__pixeloa__generateItemQr",
"mcp__pixeloa__addCatalogItem"
],
"deny": [
"mcp__pixeloa__deleteItem",
"mcp__pixeloa__unpublishItem",
"mcp__pixeloa__deleteCatalog",
"mcp__pixeloa__removeCatalogItem"
]
}
}Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as pixeloa. Review it before you use it.
Capability tags behind 2 rules
- mcp__pixeloa__publishItem outbound
- mcp__pixeloa__getItemLinks untrusted-input
Pin
Keep the decision: fail CI when the tools change, follow every change, or show a badge.
Pin in CI
Fail a build when the tools you reviewed are not the tools the server serves today.
curl -s "https://radar.anlyon.com/api/v1/servers/ai.pixeloa/pixeloa/verify?hash=1553f9de221c430143d4d29789eda7ed6548cfc86ff8a1cfa585553a0e40106e" | jq -e '.changed == false'Pinned hash 1553f9de221c, from Nov 11, 2026, 21:17 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/ai.pixeloa/pixeloa/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/ai.pixeloa/pixeloa)Source: MCP Registry. Synced Oct 21, 2026, 17:42 UTC.
Pixeloa MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the Pixeloa MCP server safe?
- No scan can promise that. What its definitions show: 20 tools, 11 of which can change, send or delete something, and 0 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.
Can the Pixeloa MCP server delete data?
- Yes. 4 tools can delete, cancel or revoke something: deleteItem, unpublishItem, deleteCatalog and removeCatalogItem.
Can the Pixeloa MCP server do things that cannot be undone?
- Yes. 1 tool can send, pay, publish or merge: publishItem.
What tools does the Pixeloa MCP server have?
- 20 tools: 4 destructive, 1 irreversible, 6 write, 9 read. deleteItem, unpublishItem, deleteCatalog, removeCatalogItem, publishItem, createItem, updateItem, confirmScale and 12 more.
Does the Pixeloa MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the Pixeloa MCP server send data to outside services?
- 1 tool can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): publishItem.
Has the Pixeloa MCP server changed its tools?
- Not since Anlyon Radar started tracking it. Last scanned Nov 11, 2026.