MCP server for web searching via DuckDuckGo and fetching web content as Markdown
Two tools with complete input schemas and descriptions, but descriptions are in Indonesian and lack LLM-optimized structure. Schemas are well-formed with proper types and enums. However, descriptions are generic and do not follow the pattern:tool-description guidance (WHAT, WHEN, prerequisites). No output schemas documented. Error handling is minimal. Parameter descriptions are adequate but lack actionable constraints. No tool annotations (readOnlyHint, idempotentHint). Tools are read-only and well-scoped, but descriptions need English translation and LLM-specific optimization.
Membaca konten halaman web dan mengonversinya ke format Markdown bersih.
Mencari informasi di DuckDuckGo dengan filter rentang waktu dan wilayah untuk mendapatkan informasi yang relevan dan terkini.
Descriptions are in Indonesian and lack LLM-optimized structure. Should follow WHAT/WHEN/prerequisites pattern and be in English for broader LLM compatibility.
No output schemas documented. LLMs cannot plan downstream calls or extract structured data without knowing what fields to expect.
No tool annotations (readOnlyHint, idempotentHint, destructiveHint). Both tools are read-only and idempotent, annotate them to help LLMs reason about safety and retry behavior.
Error handling returns plain text strings without recovery guidance. Should categorize errors (retryable, user-fixable, fatal) and suggest next steps.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 51 | 2026-07-28+ | v2 |
Parameter 'max_results' lacks minimum/maximum bounds in description. Should specify valid range (e.g., 1 - 100) to prevent LLMs from passing absurd values.