tdqs / dimensions
Dimensions
A tool is scored 1–5 on six weighted dimensions; a server is scored on four more, equally weighted, that judge the tool set as a whole. Every finding a scorer reports links to the page for its dimension.
Tool definition quality
Per tool. The weights sum to 100, and the weighted sum is the tool's TDQS.
| weight | dimension | question |
|---|---|---|
| 25% | Purpose Claritypurpose_clarity | Does the description state what the tool does? |
| 20% | Usage Guidelinesusage_guidelines | Does it say when to use this tool vs alternatives? |
| 20% | Behavioral Transparencybehavioral_transparency | Does it disclose behavior beyond what annotations already declare? |
| 15% | Parameter Semanticsparameter_semantics | Does it add meaning beyond what the input schema provides? |
| 10% | Conciseness & Structureconciseness_structure | Is it appropriately sized and front-loaded? |
| 10% | Contextual Completenesscontextual_completeness | Given the tool's complexity, is the description complete enough? |
Server coherence
Per server, equally weighted. Their mean is the coherence score.
| weight | dimension | question |
|---|---|---|
| 25% | Disambiguationdisambiguation | Can an agent tell the tools apart? |
| 25% | Naming Consistencynaming_consistency | Do tool names follow a predictable pattern? |
| 25% | Tool Count Appropriatenesstool_count_appropriateness | Is the surface well-scoped? |
| 25% | Completenesscompleteness | Are there gaps in the surface? |