three.ws Avatars & Agents
If you're building 3D agent experiences with Claude and need to work with glTF or GLB models, this server gives you programmatic access to three.ws's validation, inspection, and optimization pipeline. You can render 3D avatars from the three.ws platform directly, pull model metadata, check geometry and material properties, and run optimization passes without leaving your MCP workflow. It's purpose-built for the three.ws ecosystem, which combines WebGL rendering with LLM-driven agents and on-chain identity on EVM and Solana. Reach for this when you're orchestrating avatar workflows, preprocessing models before deployment, or need to verify that a GLB file will render correctly in the three.ws viewer before minting it on-chain.
- GitHub stars
- 269
- tools
- 16
- transport
- HTTP
- last updated
- Jun 12, 2026
Tools
16 toolsPublic tool metadata for what this MCP can expose to an agent.
list_my_avatarsList the authenticated user's avatars with id, name, slug, size, visibility, and (when permitted) direct `model_url`.
List the authenticated user's avatars with id, name, slug, size, visibility, and (when permitted) direct `model_url`.
No parameter schema in public metadata yet.
get_avatarFetch a single avatar by id or owner+slug; returns metadata plus a public `model_url` or short-lived signed URL for private avatars.
Fetch a single avatar by id or owner+slug; returns metadata plus a public `model_url` or short-lived signed URL for private avatars.
No parameter schema in public metadata yet.
search_public_avatarsFree-text + tag search across the public avatar gallery; useful for finding characters to render without prior knowledge of an id.
Free-text + tag search across the public avatar gallery; useful for finding characters to render without prior knowledge of an id.
No parameter schema in public metadata yet.
render_avatarProduce an HTML `<model-viewer>` snippet that renders the given avatar, with configurable background, camera orbit, poster, and AR button.
Produce an HTML `<model-viewer>` snippet that renders the given avatar, with configurable background, camera orbit, poster, and AR button.
No parameter schema in public metadata yet.
render_avatar_imageRender a stored avatar to a real PNG/JPEG/WebP image (headless three.js) with camera framing, pose preset, and ARKit-52 expression; cached.
Render a stored avatar to a real PNG/JPEG/WebP image (headless three.js) with camera framing, pose preset, and ARKit-52 expression; cached.
No parameter schema in public metadata yet.
delete_avatarSoft-delete an avatar you own. Requires the `avatars:delete` scope.
Soft-delete an avatar you own. Requires the `avatars:delete` scope.
No parameter schema in public metadata yet.
validate_modelRun the Khronos glTF-Validator against a public https GLB/glTF URL; returns a structured report of errors, warnings, infos, and hints. SSRF-hardened.
Run the Khronos glTF-Validator against a public https GLB/glTF URL; returns a structured report of errors, warnings, infos, and hints. SSRF-hardened.
No parameter schema in public metadata yet.
inspect_modelParse a GLB/glTF and return structural stats: scene/node/mesh counts, vertex and triangle totals, material and texture summaries, and extensions used. Pure inspection - no advice.
Parse a GLB/glTF and return structural stats: scene/node/mesh counts, vertex and triangle totals, material and texture summaries, and extensions used. Pure inspection - no advice.
No parameter schema in public metadata yet.
optimize_modelInspect the model and return actionable suggestions for reducing size and draw-call overhead: triangle budget, Draco/Meshopt, oversized textures, KTX2 transcoding, non-indexed primitives, redundant materials, and more.
Inspect the model and return actionable suggestions for reducing size and draw-call overhead: triangle budget, Draco/Meshopt, oversized textures, KTX2 transcoding, non-indexed primitives, redundant materials, and more.
No parameter schema in public metadata yet.
solana_agent_reputationComputed reputation summary for a Solana-registered agent: total/verified feedback counts, raw + verified-only score averages, validation pass/fail, task-acceptance, and dispute counts.
Computed reputation summary for a Solana-registered agent: total/verified feedback counts, raw + verified-only score averages, validation pass/fail, task-acceptance, and dispute counts.
No parameter schema in public metadata yet.
solana_agent_attestationsList recent on-chain attestations (feedback, validation, task offers, acceptances, disputes) about a Solana agent; each row includes verified/disputed/revoked flags.
List recent on-chain attestations (feedback, validation, task offers, acceptances, disputes) about a Solana agent; each row includes verified/disputed/revoked flags.
No parameter schema in public metadata yet.
solana_agent_passportFull discovery card for a Solana agent: identity, owner wallet, reputation summary, latest validation result, and attestation schema endpoint - the Solana equivalent of an ERC-8004 passport.
Full discovery card for a Solana agent: identity, owner wallet, reputation summary, latest validation result, and attestation schema endpoint - the Solana equivalent of an ERC-8004 passport.
No parameter schema in public metadata yet.
pumpfun_recent_claimsMost recent pump.fun GitHub social-fee claim events with full enrichment: GitHub profile, X/Twitter follower data, influencer tier, first-time-claim flag, fake-claim detection, and AI summary.
Most recent pump.fun GitHub social-fee claim events with full enrichment: GitHub profile, X/Twitter follower data, influencer tier, first-time-claim flag, fake-claim detection, and AI summary.
No parameter schema in public metadata yet.
pumpfun_recent_graduationsTokens that recently graduated from the bonding curve to PumpAMM, with creator and holder analysis.
Tokens that recently graduated from the bonding curve to PumpAMM, with creator and holder analysis.
No parameter schema in public metadata yet.
pumpfun_token_intelFull intel on a pump.fun token: graduation status, bonding-curve progress, creator profile, top holders, volume, bundle detection, and trust signals.
Full intel on a pump.fun token: graduation status, bonding-curve progress, creator profile, top holders, volume, bundle detection, and trust signals.
No parameter schema in public metadata yet.
pumpfun_creator_intelReputation profile for a pump.fun creator wallet: prior launches, graduation rate, claim activity, and behavioural trust signals.
Reputation profile for a pump.fun creator wallet: prior launches, graduation rate, claim activity, and behavioural trust signals.
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageReadme
Details
- Repository
- github.com/nirholas/three.ws
- Website
- three.ws
- Categories
- AI & LLM Tools, Design & Creative
- Transport
- HTTP
- Registry
- MCP registry, active
More from nirholas
- UCAI33
- PumpFun Claims Bot16
- Ethereum & EVM Message Signing23
- Ethereum Transaction Builder23
- Ethereum Wallet Generator23
- Ethereum Address Validator23
- Ethereum Keystore Encryption23
- GitHub to MCP Converter28
- Universal Blockchain MCP37
- DeFi Trading Agents32
- Documentation Extractor29
- MCP Registry Monitor26
- SperaxOS16
- Repository Intelligence25
- Crypto Tools Registry23
- 3D AI Agent Avatar269
- three.ws 3D Agent Tools269
- three.ws AgenC269
- three.ws Agora269
- three.ws Assistant Widget269
- three.ws Autopilot269
- three.ws Coin Clash269
- three.ws Herald269
- three.ws Knock269
Related AI & LLM Tools MCP servers
View all
Pdf Mcpjztan/pdf-mcpAgentic RAG over one PDF or a whole folder: hybrid search, selective page reads, tables, OCR.168stars
freellmpool0xzr/freellmpoolOpenAI-compatible MCP gateway cataloging 22 LLM providers, with routing, failover, and tools.146stars
Flow Nexusruvnet/flow-nexusCloud-powered AI platform with multi-agent swarms, sandboxes, and workflow automation93stars
Continuumpouyahasanamreji/continuumShared memory + orchestration for your coding agents. Local-first MCP, vector RAG.74stars
Lumenahmedeid1/lumenSelf-hostable agentic-AI LMS: catalog, RAG tutor, FSRS reviews, AI authoring, ingest.69stars
DOMShellapireno/domshellDrive Chrome with filesystem commands (ls, cd, grep, click, type). One MCP tool, multi-agent.59stars