---
name: Media Crawler Tieba
slug: media-crawler-tieba
category: Automation
description: Media Crawler Tieba collects Baidu Tieba threads and user evidence using keyword or bar discovery, exact thread detail, replies, and creator pages. Use it for community pain-point, vocabulary, objection, topic, or user research while preserving thread context.
github: "https://github.com/tsingyuai/growth-lab/tree/main/collectors/media-crawler-tieba"
language: Python
stars: 1734
forks: 150
install: "npx degit https://github.com/tsingyuai/growth-lab/tree/main/collectors/media-crawler-tieba ~/.claude/skills/media-crawler-tieba"
installs_to: ~/.claude/skills/media-crawler-tieba
source_path: collectors/media-crawler-tieba/SKILL.md
collection_size: 19
category_size: 1523
collection_url: "https://dirskills.com/collections/tsingyuai/growth-lab"
added: 2026-08-19T07:25:48.738Z
last_synced: 2026-08-19T07:25:48.738Z
canonical_url: "https://dirskills.com/skills/media-crawler-tieba"
---

# Media Crawler Tieba

Media Crawler Tieba collects Baidu Tieba threads and user evidence using keyword or bar discovery, exact thread detail, replies, and creator pages. Use it for community pain-point, vocabulary, objection, topic, or user research while preserving thread context.

**Install:**

```bash
npx degit https://github.com/tsingyuai/growth-lab/tree/main/collectors/media-crawler-tieba ~/.claude/skills/media-crawler-tieba
```

## README

# Tieba collection

Follow [media-crawler](../media-crawler/SKILL.md). Require purpose, keyword/bar/thread/user inputs, target size and time bounds, reply depth, and invoking Model Memory destination.

Use `--platform tieba --type search` for discovery. Configure bar names in `TIEBA_NAME_LIST` when the research scope is community-specific; otherwise use intent-distinct keywords. Configure exact thread IDs in `TIEBA_SPECIFIED_ID_LIST` and full user-home URLs in `TIEBA_CREATOR_URL_LIST`; use detail/creator mode.

Preserve bar name, thread title, original post, floor/reply relationships, author, publish time and canonical URL. Do not flatten replies into independent opinions without their parent context. Sample across different threads/authors and deduplicate by thread ID plus post/floor identity. Start without deep replies, enrich only relevant threads, and stop on authentication/risk control.

Return scope/bar list, exact inputs, raw/copied paths, thread/reply/user counts, coverage, time, commit, exclusions and selection rationale.