Loading ToolBench
AI Content Creation Assistant MCP - 한국 콘텐츠 크리에이터를 위한 트렌드 분석 및 콘텐츠 생성 도우미
17 tools with consistent naming (verb_noun pattern) and descriptions present. All tools have input schemas with typed parameters and descriptions. However, output schemas are not documented in the source code, critical for LLM planning. Parameter descriptions are generic and lack actionable constraints (enums, ranges, formats). No error handling guidance visible. Tool composition is sound (single responsibility), but descriptions average ~80 chars, below the 194-char baseline for production tools. Missing dependency hints and recovery guidance.
Analyze competitor content strategy and performance
Analyze influencer collaboration opportunities
Analyze news trends by category to identify emerging topics and sentiment
Analyze SEO keywords for search volume, competition, and relevance
Analyze and optimize video thumbnail for engagement
Benchmark content performance against industry standards
Create content calendar based on Korean events and trends
Generate A/B test variants for content optimization
Output schemas not documented. LLMs cannot plan downstream tool calls or extract required fields without knowing response structure. Critical for tool chaining.
Parameter descriptions lack actionable constraints. 'source' in get_korean_trends lists values but not as enum; 'limit' has no min/max bounds; 'category' in analyze_news_trends lacks enum constraint. LLMs hallucinate invalid values.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 55 | <=2025-11-25 | v2 |
Generate content ideas based on topic, platform, and trends
Generate comprehensive hashtag strategy for content
Generate a script outline for video content
Get Korean trends from multiple sources (Naver, Daum, Google, YouTube, Zum)
Get seasonal content guide for Korean market
Optimize content title and generate hashtag strategy
Predict content performance metrics based on historical data
Predict viral potential score for content
Repurpose existing content for different platforms
No error handling guidance visible in tool definitions. If a scraper fails or API returns no results, LLMs have no recovery path. Missing pattern:recovery-guide.
Tool descriptions average ~80 chars, below 194-char production baseline. Descriptions lack WHEN to use, dependencies, and prerequisites. E.g., 'Generate content ideas based on topic, platform, and trends' does not explain when to call it vs. generate_script_outline.
No tool composition guidance. Tools like generate_content_ideas and generate_script_outline likely depend on each other, but no description hints at the workflow or required inputs from prior calls.