Loading ToolBench
A sovereign, local-first MCP server for AI search crawler governance, robots.txt auditing, brand safety guardrails, and IndexNow notifications.
Server has 7 tools with basic descriptions and input schemas, but significant gaps in parameter documentation, output schema clarity, and error handling. All tools are READ_ONLY except notify_indexnow_search_engines (WRITE). Naming follows verb_noun convention (audit_, verify_, enforce_, check_, validate_) which is good. However, parameter descriptions are minimal or missing, output schemas are not formally documented, and error responses lack recovery guidance. No tool annotations (readOnlyHint/destructiveHint) present. Average tool score: 52/100.
Governance Tool: Parses robots.txt to verify access or block rules for AI search bots (GPTBot, ClaudeBot, PerplexityBot).
Indexing Tool: Fetches and validates XML sitemaps for syntax errors, URL counts, and lastmod freshness.
Crawl Tool: Traces HTTP redirect loops (301/302) to ensure clean canonical resolution.
Brand Safety Tool: Scans content strings for restricted keywords, offensive terms, or unverified claims.
Indexing Tool: Generates IndexNow API payloads to notify Bing, Yandex, and Naver of updated URLs.
Crawl Tool: Simulates Googlebot and ClaudeBot user-agent requests to inspect rendered HTML status codes.
Output schemas not documented. Tools return JSON strings but LLMs cannot infer field structure, types, or required fields. No documentation of pagination, error fields, or chaining IDs.
Parameter descriptions are minimal or missing. 'domain_or_url' lacks format guidance (does it accept 'example.com' or 'https://example.com'?). 'restricted_keywords_csv' lacks examples of valid input. LLMs cannot infer constraints.
Error handling lacks recovery guidance. Tools return generic JSON with 'status: ERROR' and 'message: str(e)' but do not tell LLM what to do next (retry? ask user? try alternative tool?). No error categorization (retryable vs fatal).
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 59 | 2026-07-28+ | v2 |
SEO Governance Tool: Audits HTML <link rel="canonical"> self-referential integrity and detects cross-domain redirects.
No tool annotations. Tools lack readOnlyHint/destructiveHint/idempotentHint metadata. notify_indexnow_search_engines is WRITE but has no destructiveHint. Clients cannot infer safety properties.
Incomplete source code. audit_xml_sitemap_index tool definition is truncated in provided source. Cannot fully assess schema or error handling for this tool.