comments-analysis
Skill7 starsChanged 5 months agoMITView on GitHub
Pull and analyze the comments on a post/video — sentiment breakdown, recurring themes/keywords, top comments, and notable questions or complaints. Use when the user says "analyze the comments", "what are people saying on this post", "comment sentiment", or pastes a post URL and wants the discussion summarized.
- Reads credentials
In this repository18 other files it gives its agents.
- bulk-data-exportSkill
- competitor-analysisSkill
- creator-analyticsSkill
- douyinSkill
- hashtag-researchSkill
- instagramSkill
- social-listeningSkill
- social-media-downloaderSkill
- tikhub-endpoint-discoverySkill
- tikhub-mcpSkill
- tikhub-onboardingSkill
- tikhub-python-sdkSkill
- tikhub-rest-apiSkill
- tiktokSkill
- trend-researchSkill
- twitter-threadsSkill
- xiaohongshuSkill
- youtubeSkill
The file
Files
Comments Analysis
Mine a single post's comment section for sentiment and themes.
Setup gate
[ -z "${TIKHUB_API_KEY:-}" ] && echo "Set TIKHUB_API_KEY first (see tikhub-onboarding)."Workflow
- Identify the post (platform + id/URL).
- Fetch comments, paginating to a target count (cap it):
- TikTok:
app/v3/fetch_video_comments(aweme_id,cursor). Douyin:app/v3/fetch_video_comments(aweme_id,cursor). - Instagram:
v2/fetch_post_comments(+fetch_comment_replies). - YouTube:
web_v2/get_video_comments(+get_video_comment_replies). - Twitter:
web/fetch_post_comments. Xiaohongshu:app_v2/get_note_comments.
- TikTok:
- Optional fast keywords: TikTok
analytics/fetch_comment_keywords(item_id) gives a comment keyword summary directly. - Analyze: sentiment breakdown, top themes with example quotes, most-liked comments, and any recurring questions/complaints.
- Deliver a summary with cited example comments.
Cost awareness
Each comment page (and reply page) is a billed call. Cap pages; warn for viral posts with huge
threads. Estimate via calculate_price.
Verification gate
- Comments fetched and tied to the right post.
- Themes/sentiment backed by quoted comments.
- Report the sample size (comments analyzed vs. total).
Red flags
- Summarizing 50 comments on a 50k-comment post as representative — disclose the sample.
- Unbounded reply pagination.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.