Loading ToolBench
Precompiled eligibility and pre-registration for primary care
ReadyPass has 9 tools with schemas and descriptions present, but quality is inconsistent. Tool names lack clear action verbs (approve_gap, resolve_document, resolve_all are vague about what 'resolve' means). Descriptions are present but brief (avg ~60 chars, below the 194-char baseline). Parameter descriptions exist but lack detail on constraints, formats, and dependencies. Output schemas are not documented, callers cannot see what fields to expect. Error handling is minimal; no recovery guidance. The codebase shows sophisticated domain logic (audit trails, rule compilation, approval workflows) but the tool interface does not expose this clarity to LLMs.
Submit a document to Azure Document Intelligence and poll until the analysis completes.
Turn a proposal into an approved, compiled, activated, audited rule.
Deterministic extraction from an ingested document.
Resolve the configured document understanding provider. Defaults to local.
Split the organiser's combined chit document into individual documents.
Record an audit entry in the append-only audit log.
Every document in the pack, for the demo and the regression suite.
Vague action verbs in tool names. 'resolve_document' and 'resolve_all' do not clearly indicate what 'resolve' means (admission decision? validation? extraction?). LLMs cannot infer intent from the name alone.
No output schemas documented. Tools return objects but callers cannot see field names, types, or structure. LLMs cannot plan downstream calls or extract required data.
Parameter descriptions lack constraint details. 'service_date' has format='date' but no description explaining what it controls or when it is required. 'ruleset' is described as 'Compiled rule set to use' but no guidance on how to obtain one.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 48 | 2026-07-28+ | v2 |
Take one presented document through to an admission decision.
Build an ExtractedDoc from the Azure Document Intelligence response.
No error handling or recovery guidance. Tools do not document what errors are possible, whether they are retryable, or what the LLM should do next.
Tool descriptions are too brief (avg ~60 chars vs 194-char baseline). 'Turn a proposal into an approved, compiled, activated, audited rule' is domain-specific jargon; LLMs need WHEN to call this, WHAT it does, and WHAT it returns in plain language.