Overte (open-source Vircadia-lineage metaverse) Model Context Protocol server for domain administration and in-world entity orchestration
Overte MCP has 9 tools with reasonable naming (verb_noun pattern) and descriptions present for all. However, critical gaps undermine quality: (1) Parameter descriptions are sparse or missing for many inputs, e.g., overte_entity_spawn's 'scale' param lacks detail on behavior when omitted; (2) Output schemas are NOT documented anywhere, LLMs cannot predict return structure; (3) Error handling is absent, no guidance on recovery or retryability; (4) No tool annotations (readOnlyHint, destructiveHint, idempotentHint) despite clear risk levels (DESTRUCTIVE, WRITE, READ_ONLY); (5) Tool descriptions average ~80 chars, below the 194-char baseline for A+ tools. Naming is solid (overte_entity_spawn, overte_entity_delete, overte_nearby_entities follow verb_noun), but lack of output documentation and error guidance are major blockers for production use.
Retrieve connected-node telemetry and settings from an Overte Domain Server.
Loop-animate an entity in place (spin or bob) for a fixed duration.
Permanently delete an in-world entity.
Spawn a virtual object, 3D GLB model, or Light in-world at the specified coordinates.
Move, resize, re-parent, or toggle an existing in-world entity.
Spawn a preset test fixture (box, cup, ball, table, chair) for gripper/manipulation testing.
Find real in-world entities near a point (default: the local user).
No output schemas documented. LLMs cannot predict return structure, field names, or types. Breaks tool chaining and forces agents to guess at response format.
Missing tool annotations (readOnlyHint, destructiveHint, idempotentHint). Tools marked DESTRUCTIVE (overte_entity_delete) or WRITE (overte_entity_spawn) lack formal hints. Agents cannot distinguish safe from unsafe operations.
No error handling or recovery guidance. Tools lack descriptions of failure modes, retryability, or actionable error messages. E.g., overte_entity_spawn does not document what happens if model_url is invalid.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 52 | 2026-07-28+ | v2 |
Get multi-step Overte operation guidance via MCP sampling.
Inject a JavaScript script to govern behavior of an in-world entity.
Parameter descriptions incomplete or missing. overte_entity_spawn's 'scale' param lacks detail on behavior when null; 'script_data' in overte_script_inject is vague ('Metadata parameters'). Ambiguous params force LLM guessing.
overte_sampling_assist violates single-responsibility principle. Tool name suggests it calls MCP sampling, but sampling is deprecated (2026-07-28 spec). Purpose unclear, is it a discovery tool or a planning assistant? Rename and clarify intent.