Single tool 'add' with minimal definition quality. The tool has a valid schema but severely lacks descriptive content and parameter documentation. Tool name is non-specific (not verb_noun convention), input parameters have no descriptions, and output schema is undocumented. This is a bare-minimum implementation that would fail code review against production patterns.
Add two numbers
Tool name 'add' lacks action verb clarity. Per Arcade patterns, tool names should follow verb_noun convention (e.g., 'calculate_sum', 'add_numbers') to help LLMs infer intent from the name alone. A bare 'add' is ambiguous, add what? Add where?
Input parameters 'a' and 'b' have null descriptions. LLMs cannot infer that 'a' and 'b' are numbers to be summed, they need explicit guidance.
Tool description 'Add two numbers' is only 16 characters. This description lacks context about WHEN to use the tool, what it returns, and any prerequisites. Descriptions under 20 chars score no higher than 20-40.
Output schema is undocumented. The tool returns a String result of the sum, but LLMs have no formal schema telling them to expect a numeric string or how to parse it. Per pattern:tool, all tools must document output schema so agents can plan downstream calls and extract data correctly.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | F | 34 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 26 | - | v1 |
No input validation or error guidance. If an LLM passes non-numeric strings or null values for 'a' or 'b', the tool will fail. No error message guides recovery. Per pattern:recovery-guide, error responses must tell the LLM what to do next.