# v2ex-topic-analyzer > Analyze V2EX topic threads by fetching the topic and replies via the V2EX API v2 and producing a structured written analysis. Use when asked to analyze a V2EX topic by ID, summarize a V2EX discussion, or extract insights from a V2EX thread; requires a V2EX Personal Access Token (V2EX_TOKEN) and the bundled fetch script. - Author: quicksandzn - Repository: quicksandznzn/claude-agent-skills-case - Version: 20260116161653 - Stars: 0 - Forks: 0 - Last Updated: 2026-02-06 - Source: https://github.com/quicksandznzn/claude-agent-skills-case - Web: https://mule.run/skillshub/@@quicksandznzn/claude-agent-skills-case~v2ex-topic-analyzer:20260116161653 --- --- name: v2ex-topic-analyzer description: Analyze V2EX topic threads by fetching the topic and replies via the V2EX API v2 and producing a structured written analysis. Use when asked to analyze a V2EX topic by ID, summarize a V2EX discussion, or extract insights from a V2EX thread; requires a V2EX Personal Access Token (V2EX_TOKEN) and the bundled fetch script. --- # V2EX Topic Analyzer ## Overview Fetch the topic and replies with the bundled script, then write a structured analysis using the framework in `references/analysis-framework.md`. ## Workflow 1. Collect inputs: - Required: `topic_id` - Optional: `max_pages` (default 1), `max_chars` to truncate long content, `max_replies` to cap replies - Confirm `V2EX_TOKEN` is set in the environment. 2. Fetch data: - Run: `python3 .claude/skills/v2ex-topic-analyzer/scripts/fetch_v2ex_bundle.py --topic-id --max-pages ` - If the bundle is too large, rerun with `--max-chars ` or `--max-replies `. 3. Analyze: - Read `references/analysis-framework.md`. - Answer each section in order with concrete evidence from the topic and replies. - If information is missing, say so explicitly. - Write in Chinese unless the user explicitly requests another language. 4. Output: - Return only the analysis body in Markdown and language must be chinese. - Do not include a top-level title, tool logs, or the raw bundle.