Loading ToolBench
Owner-controlled public content tools for MCP, WebMCP, llms.txt, and sitemaps. Exposes website content to AI agents through multiple surfaces including Model Context Protocol, llms.txt format, and WebMCP.
Corsen Context provides 9 tools with consistent naming (verb_noun pattern: search_site, get_page_content, list_content) and basic descriptions. All tools have input schemas with type definitions and parameter descriptions. However, descriptions are brief (10-50 chars), lacking LLM-optimized context about WHEN to use each tool or dependencies between them. Output schemas are not documented. Error handling guidance is absent. The WordPress plugin architecture suggests tool definitions are PHP-based, making full schema verification difficult from TypeScript examples alone.
Check if an agent has access to specific content (WordPress extension)
Retrieve the full content of a specific page by URI
Retrieve product information (WordPress WooCommerce extension)
Retrieve page sections and structure (WordPress extension)
Retrieve the site's XML sitemap
Retrieve structured data (schema.org) from a page (WordPress extension)
List content items of a specific type with pagination
Descriptions are too brief (10-50 chars) and lack LLM-optimized context. Missing WHEN to use each tool, dependencies, and prerequisites. Baseline for A+ tools is 50-200 chars with clear selection criteria.
Output schemas are not documented. LLMs cannot plan downstream calls or extract required fields without knowing what each tool returns. All 9 tools lack return type documentation.
No error handling guidance. Tools lack recovery hints (e.g., 'If page not found, try search_site() first'). Error responses should tell LLMs what to do next, not just fail.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 62 | <=2025-11-25 | v2 |
Request a callback from a human expert (WordPress extension)
Search the site content by query string
list_content and search_site lack explicit pagination guidance in descriptions. No mention of result limits, total counts, or next_cursor patterns. Baseline requires pagination documentation.
request_expert_call (WRITE operation) lacks confirmation/dry-run pattern. Irreversible operations should support confirmation before execution to prevent accidental callbacks.