01Safety
30.0 / 30
What changed in the harness
Selection accuracy 97→97, token cost up 8%, 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
28.8 / 30
03Economics
20.0 / 20
04Discoverability
9.4 / 20
Highest-impact fix
Estimated gain +11 pointsMake target tools discoverable on the first call
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps.
Description evidence
Defects and rewrites.
2 defects 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 |
|---|---|---|
introspect-schema |
no_return_description |
Retrieve the GraphQL schema of the endpoint before running queries, so you can learn the available types and fields when you do not already have the schema as a resource. Returns the schema information (types, fields, and their descriptions) in JSON. |
query-graphql |
params_unexplained no_return_description |
Execute a GraphQL query against the endpoint and return the matching result data in JSON. The query parameter is the GraphQL query string to run, and the optional variables parameter is a JSON string containing the variable values referenced by the query. |
Selection evidence
Confusable tool pairs.
0 pairs where similar names or overlapping descriptions may send an agent toward the wrong tool.
| Tool A | Tool B | Confidence | Why they collide |
|---|---|---|---|
| No confusable tool pairs were flagged in this assessment. | |||
Compare the field