Loading ToolBench
Check whether a website is visible to AI search engines (ChatGPT, Perplexity, Claude, Google AI Overviews) and generate starter fixes. Dependency-free; stdio plus a custom JSON-RPC-over-HTTP POST adapter.
Two tools with clear, actionable descriptions (194 - 250 chars, within baseline 34 - 392). Input schemas present with proper types and required fields. However, parameter descriptions are minimal or absent, 'url' lacks guidance on format/validation. No output schemas documented, forcing LLMs to infer response structure. No error handling guidance (e.g., 'If domain unreachable, try with https:// prefix'). Tool names are descriptive but lack verb-noun clarity (check_* and generate_* are acceptable but could be more specific). No tool annotations (readOnlyHint, idempotentHint) despite both being read-only operations.
Check whether a website is visible to AI search engines (ChatGPT, Perplexity, Claude, Google AI Overviews). Fetches the site + robots.txt and scores AI-crawler access, structured data (JSON-LD), title/meta, Open Graph, sitemap, and llms.txt. Returns a 0-100 score, a letter grade, and a specific fix for each gap.
Generate starter fixes to make a website visible to AI search: Organization + FAQPage JSON-LD (pre-filled from the site's real title/description) and an AI-crawler-friendly robots.txt. Paste the JSON-LD into the homepage <head> and replace robots.txt. This is the free starter version of the same-day Fix Pack.
Parameter 'url' lacks format/validation guidance. No description of accepted formats (bare domain vs https:// vs http://), character limits, or error cases (invalid TLD, unreachable host).
No output schema documented. LLMs cannot infer the structure of the 'score', 'grade', 'checks' array, or 'fix' field. Baseline: 100% of A+ tools document return types.
No error handling guidance. If a domain is unreachable, DNS fails, or robots.txt is malformed, the LLM receives no recovery hint. Baseline: error responses must guide next steps.
Tool annotations missing. Both tools are read-only and idempotent, should declare readOnlyHint: true and idempotentHint: true to guide agent planning and retry logic.
Inferred effective spec: 2026-07-28+.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 58 | 2026-07-28+ | v2 |
Tool names lack verb-noun clarity. 'check_ai_readiness' is acceptable but 'generate_ai_readiness_fixes' is vague, does it create files, return JSON, or send emails? Rename to 'generate_json_ld_fixes' or 'generate_robots_txt_fixes' for clarity.