Loading ToolBench
MCP server that exposes GoAnyAPI public APIs as tools, including domain analysis, keyword research, SERP queries, and advertising transparency data
GoAnyAPI MCP server exposes 17 tools with consistent naming (verb_noun pattern: goanyapi_*) and complete input schemas. All tools have descriptions (avg ~80 chars, within baseline 34-392). However, descriptions are generic and lack WHEN/WHY guidance. Parameter descriptions are minimal (e.g., 'Domain name, for example example.com' repeats across 5+ tools). No output schemas documented, LLMs cannot predict response structure for chaining. Error handling exists (toToolError) but lacks recovery guidance. Tool annotations present (readOnlyHint, destructiveHint, idempotentHint, openWorldHint) but openWorldHint=true on all tools is overstated, these are deterministic API calls. No pagination support visible despite tools returning lists (e.g., goanyapi_top10_serp, goanyapi_google_serp). Composition is sound (each tool does one thing), but missing chaining IDs in responses and result limits.
Query the authenticated account credit activity.
Query advertising statistics using a supported statistics action.
Query advertising transparency data by keyword, domain, or creative IDs.
Query AdSense information by domain or publisher ID.
Query backlink data for a domain.
Get Bing autocomplete suggestions for a query.
No output schemas documented. LLMs cannot predict response structure, breaking tool chaining and forcing exploratory calls. E.g., goanyapi_google_serp returns structured results but schema is invisible to the agent.
Descriptions lack WHEN/WHY context. E.g., 'Query website traffic, ranking, and traffic sources for a domain' does not explain when to call this vs goanyapi_domain_rating or when it's a prerequisite for other tools. Baseline: 50-200 chars with actionable guidance.
Parameter descriptions are repetitive and minimal. 'Domain name, for example example.com' appears in 5+ tools but does not explain format constraints (TLD required? subdomain allowed?). Baseline: 72 chars avg with format/range guidance.
Inferred effective spec: 2025-06-18+.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 62 | 2025-06-18+ | v2 |
Query structured Bing search results with pagination and localization options.
Query the authenticated account credit balance.
Query the domain rating for a domain.
Get Google autocomplete suggestions for a query.
Find Google results whose titles contain the requested text.
Query structured Google search results.
Search Google results limited to a specified website.
Query keyword difficulty and related search metrics.
Generate related keyword ideas for a keyword.
Query the top search results for a keyword.
Query website traffic, ranking, and traffic sources for a domain.
openWorldHint=true on all tools is misleading. These are deterministic API calls to GoAnyAPI, not open-world searches. Should be false or omitted. Annotation misuse confuses agent planning.
No pagination or result limits documented. Tools like goanyapi_top10_serp and goanyapi_google_serp return lists but lack limit/offset params or total_count in response. Agents cannot safely iterate large result sets.