R3
REMOTE · MCP.R3JS.ORG
Build 3D web apps: physics, audio, music, Blender import. Live edits, undo, versions, publishing.
Available components
Last scanned (7 hours ago)
Scores higher than 99% of remote servers
Reachable. Last reached 21 Nov 2026.
Category Developer Tools
How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. How we score → Why this is hard to score →
Endpoint Security97
- The endpoint's TLS certificate is valid, in date, and uses a strong key. View diagnostics → Pass
- Authorisation is enforced on tool calls, advertised via RFC 9728 protected-resource metadata. Discovery is public, which costs nothing: no tool can be invoked without a token. View diagnostics → Pass
- HTTPS is enforced; there's no plaintext access path. View diagnostics → Pass
- The HSTS (Strict-Transport-Security) header is present. View diagnostics → Pass
- DNSSEC check failed: this domain isn't protected by DNSSEC. See how to fix → View diagnostics → Fail
- The authorisation server supports Client ID Metadata Documents, the current MCP client-registration mechanism. View diagnostics → Pass
Transport & Reachability100
- Verified streamable-http transport via a live MCP handshake. View diagnostics → Pass
Schema Quality & AI Usability77
- 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).Pass
- AI-judged instruction clarity (excellent).Pass
- Context-footprint check failed: tool/resource definitions use about 27868 tokens (~238/item across 117 items; 80 tools + 37 resources), over budget; trim descriptions and params. See how to fix → Fail
- Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management98
- Stability check failed: schema churn in the 30 days we've observed: 1 tool removals, 0 breaking changes, 0 auth/transport breaks, 19 additions. See how to fix → Fail
Tool Coverage100
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 100% of tool parameters carry a description.Pass
- Structured output schemas are declared (98% of tools); any adoption earns full credit.Pass
Tool Safety97
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- 7 of 8 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "r3_publish_settings_read" implies "publish" and declares readOnlyHint instead, contradicting what its own name says it does. See how to fix → Partial
- An AI judge read all 82 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
- Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
- Supports UI / widget rendering.Pass
Checks we ran (14)
- TLS certificate
- Authorization
- HTTPS
- HSTS header
- DNSSEC
- Transport
- Schema quality
- Stability
- Tool coverage
- Injection markers
- Destructive annotations
- Judged manipulation
- MCP protocol
- Capabilities
Does not apply to this hosted endpoint (2)
- Supply Chain SecurityScored on package channels only. This is a hosted endpoint, so there is nothing to install or inspect.
- Provenance & TransparencyScored on package channels only. This is a hosted endpoint, so there is nothing to install or inspect.
The score does not measure everything. What the score does not measure →
Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.
- 21 Nov 26 0
- Server version: 3.222.0 → 3.223.0 functional
- Server version: 3.221.0 → 3.222.0 functional
- Server version: 3.220.0 → 3.221.0 functional
- 20 Nov 26 0
- Server version: 3.219.0 → 3.220.0 functional
- Server version: 3.218.0 → 3.219.0 functional
- 19 Nov 26 0
- New resource “HOWTO_TEXT.md” functional
- Server version: 3.217.0 → 3.218.0 functional
- Server version: 3.216.0 → 3.217.0 functional
- Server version: 3.215.0 → 3.216.0 functional
- Server version: 3.214.0 → 3.215.0 functional
- Server version: 3.213.0 → 3.214.0 functional
- Server version: 3.212.0 → 3.213.0 functional
- Server version: 3.210.0 → 3.212.0 functional
- Server version: 3.208.0 → 3.210.0 functional
- 18 Nov 26 0
- Server version: 3.207.0 → 3.208.0 functional
- Server version: 3.206.0 → 3.207.0 functional
- Server version: 3.205.0 → 3.206.0 functional
- Server version: 3.204.0 → 3.205.0 functional
- Server version: 3.202.0 → 3.204.0 functional
- Server version: 3.200.0 → 3.202.0 functional
- Server version: 3.199.1 → 3.200.0 functional
- 17 Nov 26 0
- Server version: 3.198.1 → 3.199.1 functional
- Server version: 3.197.0 → 3.198.1 functional
- Server version: 3.196.0 → 3.197.0 functional
- 16 Nov 26 0
- Resource “HOWTO_GAMES.md” was removed ▼ functional
- Server version: 3.195.0 → 3.196.0 functional
- Server version: 3.194.0 → 3.195.0 functional
- 15 Nov 26 0
- Server version: 3.193.0 → 3.194.0 functional
- Server version: 3.192.0 → 3.193.0 functional
- Server version: 3.191.0 → 3.192.0 functional
- 14 Nov 26 0
- New resource “HOWTO_GAMES.md” functional
- Server version: 3.190.0 → 3.191.0 functional
- Server version: 3.188.0 → 3.190.0 functional
- Server version: 3.186.0 → 3.188.0 functional
Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.
Captured 21 Nov 2026 · Probed https://mcp.r3js.org/mcp
TLS valid
Negotiated TLS 1.3 with TLS_AES_128_GCM_SHA256 .
| Subject | Issuer | Valid from | Valid until | Key | Signature | Serial |
|---|---|---|---|---|---|---|
| CN=mcp.r3js.org | CN=YE2,O=Let's Encrypt,C=US | 26 Oct 2026 | 24 Jan 2027 | ECDSA 256 | ECDSA-SHA384 | 67f99411823edf75d7cbcba072a4f7d33c6 |
| SANs: mcp.r3js.org | ||||||
| CN=YE2,O=Let's Encrypt,C=US (CA) | CN=Root YE,O=ISRG,C=US | 3 Sept 2025 | 2 Sept 2028 | ECDSA 384 | ECDSA-SHA384 | 4df3b15dd6c0784c507cd37b58e6f115 |
| CN=Root YE,O=ISRG,C=US (CA) | CN=ISRG Root X2,O=Internet Security Research Group,C=US | 13 May 2026 | 2 Sept 2032 | ECDSA 384 | ECDSA-SHA384 | 872165fc34b6e5fba8add5b3705fb53a |
| CN=ISRG Root X2,O=Internet Security Research Group,C=US (CA) | CN=ISRG Root X1,O=Internet Security Research Group,C=US | 13 May 2026 | 2 Sept 2032 | ECDSA 384 | SHA256-RSA | 6c8f1dc727c7117f7baf853ac980f9cd |
Background: What to check on a remote MCP endpoint →
DNSSEC insecure
Validation of mcp.r3js.org. — Not signed
| Zone | DS | Keys | Algorithms | Outcome |
|---|---|---|---|---|
| . | Trust anchor | 20326, 38696 | 8, 8 | Verified |
| org. | DS published | 725 | 8 | Verified |
| r3js.org. | No DS | Unsigned (proven) parent-signed NSEC/NSEC3 proves an unsigned delegation |
Authentication Enforced and verified
The endpoint asked for a token and published valid RFC 9728 metadata describing how to get one.
| Result | Enforced and verified |
|---|---|
| Enforced | On tool calls |
| HTTP status | 200 |
WWW-Authenticate challenge Bearer realm="r3", resource_metadata="https://mcp.r3js.org/.well-known/oauth-protected-resource/mcp", scope="project:read project:write code:write editor:dispose editor:run project:delete:created asset:read asset:write publish:preview support:write roadmap:suggest roadmap:vote"
Bearer realm="r3", resource_metadata="https://mcp.r3js.org/.well-known/oauth-protected-resource/mcp", scope="project:read project:write code:write editor:dispose editor:run project:delete:created asset:read asset:write publish:preview support:write roadmap:suggest roadmap:vote" | Header | Value |
|---|---|
| strict-transport-security | max-age=63072000; includeSubDomains; preload |
| x-content-type-options | nosniff |
| x-frame-options | DENY |
Protected resource metadata
| Document | https://mcp.r3js.org/.well-known/oauth-protected-resource/mcp |
|---|---|
| Retrieved | Yes |
| Resource | https://mcp.r3js.org/mcp |
| Authorisation server | https://mcp.r3js.org |
Background: How OAuth 2.1 works in the 2026 MCP spec →
Transports 2 probes
| Transport | URL | Outcome | Status | Location |
|---|---|---|---|---|
| streamable-http | https://mcp.r3js.org/mcp | Verified | 200 | |
| http (plaintext) | http://mcp.r3js.org/mcp | HTTPS enforced | 301 | https://mcp.r3js.org/mcp |
The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →
r3_sound_create Make a sound effect or seamless loop from a recipe ~415
Make a sound (a 48 kHz WAV in the account's audio files) from a recipe that R3 renders: layers of synth / fm / noise / sample sources, each with an envelope and effects (filter, distortion, bitcrush, delay, reverb, compressor, tremolo), and a master stage that can make a seamless LOOP. Nothing is uploaded and it is free (60 sounds an hour for each agent key or app connection). Start from a preset (laser, coin, jump, explosion, click, whoosh, hit, power-up, helicopter-rotor, engine-hum) with optional overrides {seed, duration, channels}, or pass a full `recipe`: r3_docs_read("SOUND_RECIPES.md") has the format, every preset's recipe and worked examples. An invalid recipe is refused with the path of the problem. You cannot hear it: the answer gives numbers (peak / RMS dBFS, clipping, silent tail, the loop seam check, warnings), the url an Audio component takes and a listen link for the user. Pass `attachTo` {projectId} to also add an Audio component playing it in that project (a loop loops and plays when the project runs; a one-shot is played from code). Names are unique per account.
| Name | Type | Req | Description |
|---|---|---|---|
| attachTo | object | – | Also add an Audio component playing it: {projectId, name?, volume?, playOnStart?}. |
| name | string | yes | What the owner sees in their audio files, e.g. "Helicopter rotor". Unique in the account. |
| overrides | object | – | With a preset: {seed, duration, channels} (seed = another variation). |
| preset | string | – | A preset: laser, coin, jump, explosion, click, whoosh, hit, power-up, helicopter-rotor, engine-hum. |
| recipe | object | – | Or a full recipe (format "r3-sound-recipe", version 1). |
| Name | Type | Req | Description |
|---|---|---|---|
| answer | string | yes | This answer in words: what was done, what R3 measured, decided or refused, and what to do next. The same text as the result's text block. Read it: the fields beside it are the data, this is the guida… |
| attached | – | yes | – |
| audioId | string | yes | – |
| durationMs | number | yes | – |
| listenUrl | string | yes | – |
| loop | boolean | yes | – |
| name | string | yes | – |
| seamless | boolean|null | yes | – |
| stats | object | yes | – |
| url | string | yes | – |
| warnings | array | yes | – |
No examples provided.
r3_sound_read Read a sound's recipe ~102
Read the RECIPE a sound was made from (the editable source), with its name, url, duration, whether it loops and the sound it was edited from. The account's own sounds, or anyone's public one; an uploaded file has no recipe. To change a sound, r3_sound_update makes a new version of it.
| Name | Type | Req | Description |
|---|---|---|---|
| audioId | string | yes | The sound's id (r3_sound_create or r3_audio_list answer it). |
| Name | Type | Req | Description |
|---|---|---|---|
| answer | string | yes | This answer in words: what was done, what R3 measured, decided or refused, and what to do next. The same text as the result's text block. Read it: the fields beside it are the data, this is the guida… |
| audioId | string | yes | – |
| durationMs | number|null | yes | – |
| isOwner | boolean | yes | – |
| isPublic | boolean | yes | – |
| loop | boolean | yes | – |
| name | string | yes | – |
| previousId | string|null | yes | – |
| recipe | object | yes | – |
| url | string | yes | – |
No examples provided.
r3_sound_update Change a sound: make a new version of it ~274
Change a sound the account owns: overrides {seed, duration, channels} applied to its own recipe, or a whole new `recipe` (read the current one with r3_sound_read, change it, pass it here; r3_docs_read("SOUND_RECIPES.md") has the format). The sound itself never changes: the change is saved as a new version under the same name (the user's audio list shows the latest), with a new url, and every project and release using an older version keeps it. Give `name` only to make a separate copy under that name instead. The same recipe as the latest version saves nothing. The answer names the user's projects still on an older version (r3_component_set moves an Audio component's url). An invalid change is refused with the path of the problem. Counts toward the 60 sounds an hour.
| Name | Type | Req | Description |
|---|---|---|---|
| audioId | string | yes | The sound to change (the account's own). |
| name | string | – | Only to make a separate COPY under this name. Default: a new version of the same sound, same name. |
| overrides | object | – | Changes to its recipe: {seed, duration, channels}. |
| recipe | object | – | Or a whole new recipe (format "r3-sound-recipe", version 1). |
| Name | Type | Req | Description |
|---|---|---|---|
| answer | string | yes | This answer in words: what was done, what R3 measured, decided or refused, and what to do next. The same text as the result's text block. Read it: the fields beside it are the data, this is the guida… |
| audioId | string | yes | – |
| durationMs | number | yes | – |
| listenUrl | string | yes | – |
| loop | boolean | yes | – |
| name | string | yes | – |
| previousId | string | yes | – |
| previousInUse | – | yes | – |
| seamless | boolean|null | yes | – |
| stats | object | yes | – |
| unchanged | boolean | yes | – |
| url | string | yes | – |
| version | number | yes | – |
| warnings | array | yes | – |
No examples provided.
r3_terrain_height Read the ground height of a terrain ~314
How high the ground of a heightfield terrain is at the world (x, z) positions you ask, exactly as physics and the drawn terrain have it, plus the terrain's lowest and highest point, centre, size and footprint. Use it to place something ON a terrain: an object stands on the ground at (x, z) when its lowest point is at the returned y (a vehicle: its wheels' bottom); the returned ground normal says how steep it is there. One call takes up to 100 points; with none it describes the terrain only. `terrain` is the terrain's Mesh, GeometryHeightfield or ColliderHeightfield (entityRef or id); leave it out when the project has one terrain. It reads the user's open R3 editor when it has the project (the terrain as on screen, a heightmap image included), else the saved project. A terrain shaped by a heightmap IMAGE needs the project open in R3 (r3_editor_open), because only a browser decodes the image. Read-only: nothing is changed.
| Name | Type | Req | Description |
|---|---|---|---|
| points | array | – | World positions to sample, e.g. [{"x": 0, "z": 12.5}]. Up to 100. |
| projectId | string | yes | The project, from r3_project_list or r3_project_create. |
| terrain | string | – | The terrain: its Mesh, GeometryHeightfield or ColliderHeightfield, by entityRef or id. Optional when the project has one terrain. |
| Name | Type | Req | Description |
|---|---|---|---|
| answer | string | yes | This answer in words: what was done, what R3 measured, decided or refused, and what to do next. The same text as the result's text block. Read it: the fields beside it are the data, this is the guida… |
| editor | – | yes | – |
| points | array | yes | – |
| projectId | string | yes | – |
| projectName | string|null | yes | – |
| source | string | yes | – |
| terrain | object | yes | – |
No examples provided.
r3_texture_apply Put a texture set on a material ~599
Wire a texture set to a material in a project: the colour map to `map`, the normal map to `normalMap`, the ambient-occlusion / roughness / metalness map to `aoMap` + `roughnessMap` + `metalnessMap` (for a set without metal its blue channel is 0, so a metal map left by an earlier set is replaced and the material's own metalness is kept), the SAME repeat on every map, the material colour white, roughness 1 and metalness 1 for a metallic set (THREE multiplies the colour map by the colour and the roughness map by the roughness: a grey #808080 colour would show the texture at half brightness, the default roughness 0.5 would halve every roughness; the colour's alpha, the opacity, is kept). A Color the material shares with another component is left as it is, and the material gets its own new white Color; the answer says which. Creates one Texture per map, or reuses a Texture of the project already on that image. MaterialStandard / Physical get every map; Lambert / Phong / Toon get map, normalMap and aoMap; Basic is refused (unlit). `repeat` is how often the tile repeats across the surface (both axes); or give `worldSize` in metres: a number for a square surface (1000 for a 1 km terrain with 1 m gravel), or {x, y} for its two sides ({"x": 2, "y": 40} for a 2 m × 40 m path), giving repeatX = x / the set's tile size and repeatY = y / it, so a long strip does not stretch the texture. Give neither and the mesh using the material is measured (its own geometry size times its scale, as r3_component_bounds measures; the two largest sides). Kept at once in the project's live copy: every R3 tab with the project open shows it.
| Name | Type | Req | Description |
|---|---|---|---|
| material | string | yes | The material to wire: its entityRef or id (r3_project_load), e.g. "material_ground". |
| projectId | string | yes | The project (r3_project_list). |
| repeat | number | – | How often the texture repeats across the surface, on both axes and every map. |
| setId | string | yes | The texture set (r3_texture_create answers it). |
| worldSize | – | – | Or: the surface size in metres, a number (square) or {x, y} (its two sides, e.g. {"x": 2, "y": 40} for a 2 m × 40 m path); repeatX = x / the set's tile size (1 m for the presets), repeatY = y / it. N… |
| Name | Type | Req | Description |
|---|---|---|---|
| answer | string | yes | This answer in words: what was done, what R3 measured, decided or refused, and what to do next. The same text as the result's text block. Read it: the fields beside it are the data, this is the guida… |
| colour | object | yes | – |
| material | object | yes | – |
| measured | – | yes | – |
| projectId | string | yes | – |
| repeat | number|null | yes | – |
| repeatX | number | yes | – |
| repeatY | number | yes | – |
| scalars | object | yes | – |
| seq | – | – | – |
| textures | array | yes | – |
No examples provided.
r3_texture_create Make a tileable texture set (colour, normal, roughness) from a recipe ~615
Make a matching, seamlessly tiling texture SET for a material: a colour map, a normal map and a packed ambient-occlusion / roughness / metalness map, rendered by R3 from a procedural recipe (seeded noise, stones, bricks …). Nothing is uploaded and it is free. The maps are stored as images in the account (the owner sees them in the Images browser as "<name> · albedo.png" …). Start from a preset (gravel, sand, concrete, bricks, brushedMetal) with optional controls: seed (another variation), scale (an integer: finer pattern), colours (the ramp's "#rrggbb" stops, exactly as many colours as the preset has: gravel 3, sand 2, concrete 3, bricks 3, brushedMetal 3), contrast, bumpiness (normal strength), roughness {min, max} (from the smooth parts to the rough parts), metalness (0 = not metal, 1 = metal; metal looks dark without an environment map in the scene), size (256 / 512 / 1024 / 2048, default 1024), tileWorldSize (metres one tile covers, default 1), height (also store a 16-bit height map). Or pass a full `recipe` (r3_docs_read("TEXTURE_RECIPES.md") has the format and every preset's recipe). An invalid recipe is refused with the path of the problem. Pass `apply` {projectId, material, repeat | worldSize (a number, or {x, y} for a long strip)} to put it on a material in the same call (as r3_texture_apply). You cannot see the texture: the answer gives numbers (mean colour, brightness range, height spread, seam check) and the colour map's url; r3_editor_capture shows it in an open editor. Names are unique per account. Each call makes a NEW set; reuse a set with r3_texture_apply.
| Name | Type | Req | Description |
|---|---|---|---|
| apply | object | – | Also wire it to a material: {projectId, material, repeat | worldSize}. |
| controls | object | – | With a preset: {seed, scale, colours, contrast, bumpiness, roughness: {min, max}, metalness, size, tileWorldSize, height}. colours: exactly the preset's number of "#rrggbb" stops (gravel 3, sand 2, c… |
| name | string | yes | What the owner sees, e.g. "Gravel path". Unique in the account. |
| preset | string | – | A preset: gravel, sand, concrete, bricks, brushedMetal. |
| recipe | object | – | Or a full recipe (format "r3-texture-recipe", version 1). |
| size | integer | – | 256, 512, 1024 (default) or 2048 pixels square. |
| Name | Type | Req | Description |
|---|---|---|---|
| answer | string | yes | This answer in words: what was done, what R3 measured, decided or refused, and what to do next. The same text as the result's text block. Read it: the fields beside it are the data, this is the guida… |
| applied | – | yes | – |
| maps | object | yes | – |
| name | string | yes | – |
| setId | string | yes | – |
| size | number | yes | – |
| stats | object | yes | – |
| tileWorldSize | number | yes | – |
| warnings | array | yes | – |
No examples provided.
r3_texture_read Read a texture set and the recipe it was made from ~97
Read one texture set: its name, size, tile size, the three map urls, the numbers its render measured, the sets it was edited from, and the full RECIPE (the editable source). The account's own sets, or anyone's public one. To change a set, r3_texture_update makes a new version of it.
| Name | Type | Req | Description |
|---|---|---|---|
| setId | string | yes | The set id (r3_texture_create answers it). |
| Name | Type | Req | Description |
|---|---|---|---|
| answer | string | yes | This answer in words: what was done, what R3 measured, decided or refused, and what to do next. The same text as the result's text block. Read it: the fields beside it are the data, this is the guida… |
| set | object | yes | – |
No examples provided.
r3_texture_update Change a texture set: make a new version of it ~350
Change a texture set the account owns: its look (controls: seed, scale, colours, contrast, bumpiness, roughness {min, max}, metalness, size, tileWorldSize, height, applied to the set's own recipe; `scale` multiplies its pattern, so 2 is twice as fine as now) or a whole new `recipe` (read the current one with r3_texture_read, change it, pass it here). The set itself never changes: the change is saved as a new version under the same name (the user's lists show the latest), with new maps, and every project and release that uses an older version's maps keeps them. Give `name` only to make a separate copy under that name instead. The same recipe as the latest version saves nothing. The answer names the user's projects still on an older version; put the new set on a material with r3_texture_apply when the user wants it. An invalid change is refused with the path of the problem. Counts as one render toward your hourly texture budget.
| Name | Type | Req | Description |
|---|---|---|---|
| controls | object | – | Changes to its recipe: {seed, scale, colours, contrast, bumpiness, roughness: {min, max}, metalness, size, tileWorldSize, height}. |
| name | string | – | Only to make a separate COPY under this name. Default: a new version of the same set, same name. |
| recipe | object | – | Or a whole new recipe (format "r3-texture-recipe", version 1). |
| setId | string | yes | The set to change (r3_texture_create / r3_texture_read answer it). The account's own. |
| Name | Type | Req | Description |
|---|---|---|---|
| answer | string | yes | This answer in words: what was done, what R3 measured, decided or refused, and what to do next. The same text as the result's text block. Read it: the fields beside it are the data, this is the guida… |
| maps | object | yes | – |
| name | string | yes | – |
| previousId | string | yes | – |
| previousInUse | – | yes | – |
| setId | string | yes | – |
| size | number | yes | – |
| stats | object | yes | – |
| tileWorldSize | number | yes | – |
| unchanged | boolean | yes | – |
| version | number | yes | – |
| warnings | array | yes | – |
No examples provided.
r3_unpublish_live Take a live project offline ~145
Take a live R3 project offline: its public live URL stops working at once, and the live-project slot it used on the plan is free again. Only when the user asked for it. Other people open a live project, so one that blocks an r3_project_delete goes offline only once the user has been asked and agreed, never as a step of the delete. Deletes nothing: the project, and any preview link, stay; r3_publish_live makes it live again. A project that is not live is refused and left as it is. Needs the publish:live right.
| Name | Type | Req | Description |
|---|---|---|---|
| projectId | string | yes | The project id (r3_project_list). |
| Name | Type | Req | Description |
|---|---|---|---|
| answer | string | yes | This answer in words: what was done, what R3 measured, decided or refused, and what to do next. The same text as the result's text block. Read it: the fields beside it are the data, this is the guida… |
| environment | string | yes | – |
| live | boolean | yes | – |
| projectId | string | yes | – |
No examples provided.
r3_whoami Who am I in R3? ~161
Report the R3 account you act for, how you are connected (an agent key, or an app connection made with Sign in with R3), what you are allowed to do, the account's plan (its name and how many live projects it allows and uses: going live needs one), and whether the R3 platform is reachable. It verifies the credential end to end and names the capabilities it carries, so it is the call that confirms the account before any other tool. For an account with no project yet it gives the first step: starters to build, and projects it can remix. Tool schemas version: 0ee7843a374b (the tool list you were given; r3_whoami returns the one served now).
Input schema present but exposes no named parameters.
| Name | Type | Req | Description |
|---|---|---|---|
| agentId | string | yes | – |
| agentName | string | yes | – |
| answer | string | yes | This answer in words: what was done, what R3 measured, decided or refused, and what to do next. The same text as the result's text block. Read it: the fields beside it are the data, this is the guida… |
| apiReachable | boolean | yes | – |
| capabilities | array | yes | – |
| connection | string | yes | – |
| editor | object | yes | – |
| environment | string | yes | – |
| label | string|null | yes | – |
| plan | object | yes | – |
| projectsOnFirstPage | number | yes | – |
| remixable | – | yes | – |
| retiredCapabilities | array | yes | – |
| starters | – | yes | – |
| toolSchemaVersion | string | yes | – |
| username | string | yes | – |
No examples provided.
How do I install the R3 MCP server?
R3 is a hosted endpoint at https://mcp.r3js.org/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.
remote · mcp.r3js.org
claude mcp add --transport http org-r3js-r3 'https://mcp.r3js.org/mcp'
What is the R3 MCP server?
R3 is an MCP server listed in the public MCP registry as org.r3js/r3. Build 3D web apps: physics, audio, music, Blender import. Live edits, undo, versions, publishing. This page covers its hosted endpoint (https://mcp.r3js.org/mcp).
Is the R3 MCP server safe to use?
R3 scores 94 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.
What tools does the R3 MCP server expose?
R3 exposes 80 tools: r3_docs_list, r3_docs_read, r3_whoami, r3_project_create, r3_project_list, and 75 more. Their descriptions and schemas cost roughly 23,934 tokens of context every time the server is loaded.
Does the R3 MCP server require authentication?
Yes. R3 asked us for credentials when we connected, so you will need to authorise it in your MCP client before it can do anything.
Is the R3 MCP server still maintained?
R3 is still listed as active in the MCP registry. We last reached this channel on 21 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.
What did VerifyMCP test on the R3 MCP server?
VerifyMCP ran 14 checks on R3, covering TLS certificate, Authorization, HTTPS, HSTS header and 10 more. Supply Chain Security and Provenance & Transparency do not apply to a hosted endpoint, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.