Loading ToolBench
chromectx defines 2 tools with basic schemas and descriptions. Both tools have input schemas with type definitions and descriptions present. However, descriptions are minimal (doctor: 67 chars, rag: 119 chars) and lack actionable context for LLM selection. The 'rag' tool has reasonable parameter descriptions, but 'doctor' has an empty input schema with no parameters documented. Output schemas are not documented anywhere in the codebase. Error handling and recovery guidance are absent. Tool composition is reasonable (single responsibility), but descriptions do not explain WHEN to use each tool or what downstream actions they enable.
Inspect the local runtime and report macOS-specific readiness.
Search open Chrome tabs and return page-grouped results with full markdown content for RAG retrieval.
doctor tool has empty input schema (no parameters) but description does not explain what it inspects or returns. LLM cannot determine what data it provides or when to call it.
No output schemas documented for either tool. LLMs cannot plan downstream calls or extract required fields (e.g., does rag return tab IDs for follow-up actions?).
rag tool description lacks guidance on when to use it vs. other search/retrieval patterns. Does not explain that it searches OPEN CHROME TABS only (not history or bookmarks).
No error handling or recovery guidance. If doctor fails (Chrome not installed), or rag finds no results, LLM receives no actionable next steps.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 48 | <=2025-11-25 | v2 |
rag 'top' parameter lacks bounds. Description says 'Maximum number of page results' but does not specify min/max (e.g., 1 - 100). LLM could pass 0 or 10000.