01Safety
0.0 / 30
What changed in the harness
Selection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
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
28.7 / 30
03Economics
19.8 / 20
04Discoverability
20.0 / 20
Highest-impact fix
Estimated gain +30 pointsAdd explicit identity and permission preflight tools
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions.
Description evidence
Defects and rewrites.
0 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 |
|---|---|---|
| No description defects were flagged in this assessment. | ||
Selection evidence
Confusable tool pairs.
6 pairs where similar names or overlapping descriptions may send an agent toward the wrong tool.
| Tool A | Tool B | Confidence | Why they collide |
|---|---|---|---|
generate_diagram_from_text |
generate_diagram_from_ascii |
medium | A user pasting a box-drawing/flow layout as text and asking to 'make a proper diagram of this' could be routed to the general text tool, which also accepts arbitrary string content, instead of the ASCII-specific converter. |
generate_diagram_from_text |
generate_diagram_from_json |
medium | A task like 'create a diagram from this JSON config/schema' might be handled by the general text tool because it accepts any string content and is the default entry point for 'make a diagram', skipping the JSON-specific tool. |
generate_diagram_from_text |
generate_diagram_from_mermaid |
medium | A user asking to 'render/diagram this mermaid code' could be routed to the general text tool instead of the mermaid-specific converter, since both accept text content and the text tool is the generic diagram entry point. |
generate_diagram_from_ascii |
generate_diagram_from_mermaid |
medium | Both tools convert an existing text-based diagram specification, so a task phrased as 'convert this text diagram/code' is ambiguous about whether the input is plain ASCII layout or Mermaid syntax. |
generate_diagram_from_ascii |
generate_diagram_from_image |
low | A vague request like 'convert this diagram I drew' is ambiguous only in the rare case where the input format (text art vs a photo/screenshot of a sketch) is unclear from the task wording. |
generate_diagram_from_json |
generate_diagram_from_mermaid |
low | Structured-data tasks could be phrased ambiguously (e.g., a schema provided as JSON-like text vs an erDiagram in Mermaid), though the descriptions distinguish the two fairly clearly. |
Compare the field