Loading ToolBench
MCP-enabled CLI for sitemap health checks, technical SEO, CI reports, and AI agent readiness audits
Sitepulse has well-structured tool definitions with complete input schemas and clear descriptions. All 9 tools have descriptions (avg ~120 chars) and detailed parameter schemas with types, enums, and defaults. However, output schemas are not documented, LLMs cannot predict what fields to expect from results. Export tools accept generic 'results' and 'report' objects without specifying their structure. Error handling is minimal; no recovery guidance or actionable error messages visible. Tool naming is clear (verb_noun pattern), but composition could be tighter: 5 export tools do similar work (CSV/JSON/HTML/JUnit/SARIF) and could be unified with a format parameter. No tool annotations (readOnlyHint/destructiveHint) despite clear risk classifications in metadata.
Run an agent readiness audit for the sitemap host, checking for robots.txt, llms.txt, llms-full.txt, homepage accessibility, metadata, structured data, semantic HTML, and AI bot access rules
Check all URLs discovered from a sitemap.xml URL for HTTP status, redirects, and optional metadata extraction
Export agent readiness audit report as HTML file
Export agent readiness audit report as JSON file
Export URL check results as CSV file
Export URL check results as HTML report with summary, results table, and slowest URLs
Export URL check results as JSON file
Output schemas not documented. Export tools accept generic 'results' and 'report' objects without specifying their structure. LLMs cannot predict what fields to expect or plan downstream operations.
Five export tools (CSV/JSON/HTML/JUnit/SARIF) perform similar work with identical input structure. Should be unified into a single 'export_results' tool with a 'format' enum parameter to reduce cognitive load and improve composability.
No tool annotations (readOnlyHint/destructiveHint/idempotentHint) despite clear risk classifications in metadata. Export tools are marked WRITE but lack destructiveHint annotation; check tools are READ_ONLY but lack readOnlyHint.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 46 | 2026-07-28+ | v2 |
Export URL check results as JUnit XML for CI systems
Export URL check findings as SARIF for code scanning systems
No error handling or recovery guidance visible. Tools lack actionable error messages (e.g., 'Invalid sitemap URL: must be valid HTTP(S) URL' or 'Timeout fetching sitemap. Try increasing timeout parameter or checking network connectivity.').
Export tools accept 'results' and 'report' as generic array/object types without documenting expected structure. LLMs cannot validate input or understand what fields are required.