01Safety
30.0 / 30
What changed in the harness
Selection accuracy 100→100, token cost down 0%, unconfirmed writes 100%→100%.
Category breakdown
Where the score comes from.
Earned points across the four signals Gradable measures. Safety and Legibility are scored out of 30; Economics and Discoverability are scored out of 20.
02Legibility
29.1 / 30
03Economics
16.7 / 20
04Discoverability
19.0 / 20
Highest-impact fix
Estimated gain +3 pointsReduce the tool-definition and task token payload
Consolidate overlapping tools and trim repetitive schema text while preserving behavior, parameter, and return semantics.
Description evidence
Defects and rewrites.
1 defect found across the exposed tool descriptions. Suggested rewrites make purpose, inputs, boundaries, and returns easier for an agent to understand.
| Tool | Defect types | Suggested rewrite |
|---|---|---|
list_kavel_models |
name_restates_behavior no_return_description |
Retrieve Kavel's catalog of AI models used across its image and video generators, returning for each model its name, what it excels at, and whether a free tier is available — use this to help a user choose or compare models before picking a tool. |
Selection evidence
Confusable tool pairs.
2 pairs where similar names or overlapping descriptions may send an agent toward the wrong tool.
| Tool A | Tool B | Confidence | Why they collide |
|---|---|---|---|
build_kavel_prompt |
open_in_kavel |
medium | Both take a tool slug and produce prompt/URL for generating an effect; a request like 'give me the prompt to make a figurine' could be satisfied by either since open_in_kavel also returns a tuned prompt embedded in the URL. |
list_kavel_tools |
list_kavel_models |
low | Both are discovery/listing calls about Kavel's AI capabilities, but one lists effect tools (hairstyle changer, figurine) and the other lists underlying models; a vague 'what can Kavel do' query could momentarily be routed to either. |
Compare the field