Skip to content

AXIS Avatar Foundry MCP server

com.jonathanarvay/axis-avatar-foundryjonathanarvay.comstreamable-httpv0.4.0Scanned

3D avatar/asset foundry: text/image -> rigged, validated, engine-ready GLB via x402.

14 tools: 9 change things and 5 only read.

Tools: 0 delete, 0 irreversible, 9 write, 5 read.
  • Reads private data · 2
  • Reads text others write · 2
  • Can send data out · 1
  • 3 findings

Tools listed without sign-in. Remote posture

With Anlyon

9 tools here change things. Govern them.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

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 AXIS Avatar Foundry MCP server can do

The AXIS Avatar Foundry MCP server is published by jonathanarvay.com, a domain the MCP Registry verified. It has 14 tools. It can create or change data (9 tools: axis_process, post_process_mesh, axis_export, axis_manifest_verify and 5 more), but cannot delete data. 5 tools only read.

2 tools read private data (axis_inspect and axis_compare); 2 tools bring in text other people can write (generate_avatar_from_text and generate_avatar_from_image); 1 tool can send data out of your system (post_process_mesh). On its own it has all three parts of a prompt-injection exfiltration path: private data, text others write, and a way out.

The fixed checks matched 3 findings (annotation-conflict). Each one comes with the text it matched.

Its remote endpoint lists its tools without sign-in.

Last scanned Nov 13, 2026, version 0.4.0.

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.

14 tools

Changes things 9Claude Code: ask
  • axis_exportExport an avatar to a target engine platform with validation and HMAC-signed manifest.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 read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Export an avatar to a target engine platform with validation and HMAC-signed manifest. [Paid: $0.50 USDC via x402 for unauthenticated calls on the hosted /mcp endpoint; settled only on a successful (non-error) result.]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • axis_manifest_verifyVerify an export manifest's integrity and optionally check content hash.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 read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Verify an export manifest's integrity and optionally check content hash. [Paid: $0.10 USDC via x402 for unauthenticated calls on the hosted /mcp endpoint; settled only on a successful (non-error) result.]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • axis_processRun the full AXIS avatar pipeline (classify → normalize → rig → skin → validate).1 finding

    Why it is in this group

    • The description's first verb is "run": a word for a change.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    Findings

    • Warn

      The server marks "axis_process" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

      Matched text

      annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "run": a word for a change.

      What this check looks for
    Description, as the server wrote it
    Run the full AXIS avatar pipeline (classify → normalize → rig → skin → validate). Returns contract summary. [Paid: $1.50 USDC via x402 for unauthenticated calls on the hosted /mcp endpoint; settled only on a successful (non-error) result.]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • generate_avatar_from_imagePAID (x402, 6 USDC): generate a production 3D avatar from a reference photo (base64) — xAI A-pose normalize → TRELLIS.2 GPU mesh → full canonical post-process.Reads text others writeNo findings

    Why it is in this group

    • The name "generate_avatar_from_image" says "generate": a word for a change.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    PAID (x402, 6 USDC): generate a production 3D avatar from a reference photo (base64) — xAI A-pose normalize → TRELLIS.2 GPU mesh → full canonical post-process. Payment settles ONLY when generation completes; a failed generation is never charged. Returns a job_id — poll with the free get_generation_status tool (typical runtime 5-15 minutes). Pass idempotency_key when a retry is possible (timeout/5xx) — replaying the same key returns the original job_id instead of a second GPU job/charge. [Paid: $6.00 USDC via x402 for unauthenticated calls on the hosted /mcp endpoint; DEFERRED settlement — charged only when the generation job completes, a failed job is never charged. Poll get_generation_status (free).]

    Annotations the server declares

    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
  • generate_avatar_from_textPAID (x402, 5 USDC): generate a production 3D avatar from a text prompt — xAI concept image → TRELLIS.2 GPU mesh → full canonical post-process (repair/validate/export).Reads text others writeNo findings

    Why it is in this group

    • The name "generate_avatar_from_text" says "generate": a word for a change.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    PAID (x402, 5 USDC): generate a production 3D avatar from a text prompt — xAI concept image → TRELLIS.2 GPU mesh → full canonical post-process (repair/validate/export). Payment settles ONLY when generation completes; a failed generation is never charged. Returns a job_id — poll with the free get_generation_status tool (typical runtime 5-15 minutes). Pass idempotency_key when a retry is possible (timeout/5xx) — replaying the same key returns the original job_id instead of a second GPU job/charge. [Paid: $5.00 USDC via x402 for unauthenticated calls on the hosted /mcp endpoint; DEFERRED settlement — charged only when the generation job completes, a failed job is never charged. Poll get_generation_status (free).]

    Annotations the server declares

    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
  • post_process_meshPost-process a RAW generated mesh (normalize → mesh-repair → rig/skin → validate → export) and return the finished, validated GLB as base64.Can send data out1 finding

    Why it is in this group

    • The name "post_process_mesh" says "post": a word for a change.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • outboundThe name "post_process_mesh" says "post": it sends or publishes something to other people.

    Findings

    • Warn

      The server marks "post_process_mesh" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

      Matched text

      annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "post_process_mesh" says "post": a word for a change.

      What this check looks for
    Description, as the server wrote it
    Post-process a RAW generated mesh (normalize → mesh-repair → rig/skin → validate → export) and return the finished, validated GLB as base64. The inline counterpart of axis_process (which needs a server file path) — the generation post-process step, callable by an LLM/agent that holds the mesh bytes (e.g. fresh TRELLIS output) or a URL to them. Provide the mesh EXACTLY one of two ways: inline as base64 (glb_base64) or as an HTTPS URL (mesh_url) — the URL's host must be on this server's operator-configured allowlist (off by default; an unconfigured allowlist refuses every mesh_url call, it is never open to an arbitrary host). Providing both or neither is a real input error. [Paid: $1.50 USDC via x402 for unauthenticated calls on the hosted /mcp endpoint; settled only on a successful (non-error) result.]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • repair_meshRepair a mesh into AXIS canonical frame (scale/axis/degenerates/normals) and return repair metrics.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 read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Repair a mesh into AXIS canonical frame (scale/axis/degenerates/normals) and return repair metrics. [Paid: $0.75 USDC via x402 for unauthenticated calls on the hosted /mcp endpoint; settled only on a successful (non-error) result.]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • retarget_animationRetarget a built-in animation clip to a platform mapping and report coverage + deterministic hashes.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 read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Retarget a built-in animation clip to a platform mapping and report coverage + deterministic hashes. [Paid: $0.75 USDC via x402 for unauthenticated calls on the hosted /mcp endpoint; settled only on a successful (non-error) result.]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • roblox_compliance_checkRun Roblox R15 compliance checks against a mesh using canonical pipeline outputs.1 finding

    Why it is in this group

    • The description's first verb is "run": a word for a change.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    Findings

    • Warn

      The server marks "roblox_compliance_check" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

      Matched text

      annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "run": a word for a change.

      What this check looks for
    Description, as the server wrote it
    Run Roblox R15 compliance checks against a mesh using canonical pipeline outputs. [Paid: $0.25 USDC via x402 for unauthenticated calls on the hosted /mcp endpoint; settled only on a successful (non-error) result.]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
Only reads 5Show 5 read-only tools
  • axis_compareCompare two avatar contracts.Reads private dataNo findings

    Why it is in this group

    • The name "axis_compare" says "compare": a word for reading.

    What it can reach

    • private-dataThe input names a file (input.file_a), and the tool reads: it reads files.
    Description, as the server wrote it
    Compare two avatar contracts. Returns similarity score and section-level diffs. [Paid: $0.10 USDC via x402 for unauthenticated calls on the hosted /mcp endpoint; settled only on a successful (non-error) result.]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • axis_inspectInspect an avatar's contract state and provenance chain (stage, bone/vertex counts, provenance step count, chain_valid + any broken links).Reads private dataNo findings

    Why it is in this group

    • The name "axis_inspect" says "inspect": a word for reading.

    What it can reach

    • private-dataThe input names a file (input.file_path), and the tool reads: it reads files.
    Description, as the server wrote it
    Inspect an avatar's contract state and provenance chain (stage, bone/vertex counts, provenance step count, chain_valid + any broken links). Does not return a quality grade — use axis_validate for pass/warn/fail rule results. [Paid: $0.10 USDC via x402 for unauthenticated calls on the hosted /mcp endpoint; settled only on a successful (non-error) result.]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • axis_list_capabilitiesList all AXIS Foundry capabilities (input formats, export platforms, validation rules).No findings

    Why it is in this group

    • The name "axis_list_capabilities" says "list": a word for reading.
    Description, as the server wrote it
    List all AXIS Foundry capabilities (input formats, export platforms, validation rules). [Free — no payment required.]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • axis_validateValidate an avatar mesh file.No findings

    Why it is in this group

    • The name "axis_validate" says "validate": a word for reading.
    Description, as the server wrote it
    Validate an avatar mesh file. Returns pass/warn/fail counts and detailed rule results. [Paid: $0.25 USDC via x402 for unauthenticated calls on the hosted /mcp endpoint; settled only on a successful (non-error) result.]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • get_generation_statusFREE: poll a generation job submitted by the paid generate tools.No findings

    Why it is in this group

    • The name "get_generation_status" says "get": a word for reading.
    Description, as the server wrote it
    FREE: poll a generation job submitted by the paid generate tools. Returns status/progress; once complete AND the deferred payment has settled, returns the download URLs. A failed generation reports charged=false. Look up by job_id, or by idempotency_key + payer if the job_id was lost. [Free — no payment required.]

    Annotations the server declares

    • 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.

4 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 1 outbound tool breaks every one.

Reads private data2
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write2
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.
deny to break every path
{
  "permissions": {
    "deny": [
      "mcp__axis-avatar-foundry__post_process_mesh"
    ]
  }
}

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.

Tools listed without sign-in

Looked at Nov 20, 2026, 18:41 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
9
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
  • WarnThe whole server

    The endpoint served tools/list without credentials, and the list has 9 tools that change things (9 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://api.avatar.jonathanarvay.com/mcp: tools/list returned 14 tools to an anonymous client, including axis_process (write), post_process_mesh (write), axis_export (write), axis_manifest_verify (write), repair_mesh (write) and 4 more

    What this check looks for

Findings

Matches of fixed checks, with the text they matched.

3 warn
  • The server marks "axis_process" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

    Matched text

    annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "run": a word for a change.

    What this check looks for
  • The server marks "post_process_mesh" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

    Matched text

    annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "post_process_mesh" says "post": a word for a change.

    What this check looks for
  • The server marks "roblox_compliance_check" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

    Matched text

    annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "run": a word for a change.

    What this check looks for

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 0.4.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http axis-avatar-foundry https://api.avatar.jonathanarvay.com/mcp
claude mcp add --transport http axis-avatar-foundry https://api.avatar.jonathanarvay.com/mcp

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.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__axis-avatar-foundry__axis_list_capabilities",
      "mcp__axis-avatar-foundry__axis_validate",
      "mcp__axis-avatar-foundry__axis_inspect",
      "mcp__axis-avatar-foundry__axis_compare",
      "mcp__axis-avatar-foundry__get_generation_status"
    ],
    "ask": [
      "mcp__axis-avatar-foundry__axis_process",
      "mcp__axis-avatar-foundry__post_process_mesh",
      "mcp__axis-avatar-foundry__axis_export",
      "mcp__axis-avatar-foundry__axis_manifest_verify",
      "mcp__axis-avatar-foundry__repair_mesh",
      "mcp__axis-avatar-foundry__retarget_animation",
      "mcp__axis-avatar-foundry__roblox_compliance_check",
      "mcp__axis-avatar-foundry__generate_avatar_from_text",
      "mcp__axis-avatar-foundry__generate_avatar_from_image"
    ],
    "deny": []
  }
}

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 axis-avatar-foundry. Review it before you use it.

Capability tags behind 5 rules
  • mcp__axis-avatar-foundry__axis_inspect private-data
  • mcp__axis-avatar-foundry__axis_compare private-data
  • mcp__axis-avatar-foundry__post_process_mesh outbound
  • mcp__axis-avatar-foundry__generate_avatar_from_text untrusted-input
  • mcp__axis-avatar-foundry__generate_avatar_from_image 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/com.jonathanarvay/axis-avatar-foundry/verify?hash=e7db653b8d4b883190875051ee50d6b0984fb50acde4cafd99b18b831e6ca407" | jq -e '.changed == false'

Pinned hash e7db653b8d4b, from Nov 13, 2026, 13:18 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/com.jonathanarvay/axis-avatar-foundry/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for com.jonathanarvay/axis-avatar-foundry
[![Anlyon Radar](https://radar.anlyon.com/badge/com.jonathanarvay/axis-avatar-foundry.svg)](https://radar.anlyon.com/servers/com.jonathanarvay/axis-avatar-foundry)

Source: MCP Registry. Synced Oct 21, 2026, 17:57 UTC.

AXIS Avatar Foundry MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the AXIS Avatar Foundry MCP server safe?

No scan can promise that. What its definitions show: 14 tools, 9 of which can change, send or delete something, and 3 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 AXIS Avatar Foundry MCP server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the AXIS Avatar Foundry MCP server do things that cannot be undone?

No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.

What tools does the AXIS Avatar Foundry MCP server have?

14 tools: 9 write, 5 read. axis_process, post_process_mesh, axis_export, axis_manifest_verify, repair_mesh, retarget_animation, roblox_compliance_check, generate_avatar_from_text and 6 more.

Does the AXIS Avatar Foundry MCP server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the AXIS Avatar Foundry 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): post_process_mesh.

Has the AXIS Avatar Foundry MCP server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Nov 13, 2026.