Bilibili Downloader
Paste the following prompt into your AI chat to install this skill:
Install @user_4ac179d5/bilibili-down into your AI assistant by following https://skillhub.cn/install/skillhub.md.
About this skill
Problem It Solves
Bilibili downloads in agent workflows are often split into separate steps: extracting a BV ID from conversation, querying parts, choosing video or audio, calling the download API, merging with ffmpeg, and handling errors. Bilibili-Down packages these steps into a stateful download flow centered on BV IDs, so upstream callers can trigger downloads with media_type and pages without re-implementing WBI signing, metadata queries, multi-part downloads, and error mapping.
How It Works and Boundaries
The skill only handles conversation-based downloads for Bilibili BV IDs and reuses local Python modules under core/ and scripts/:
- download_bilibili_media downloads the first P media
- download_bilibili_multi_p downloads all parts or selected parts via pages
- WAIT_CONFIRM is only a compatibility placeholder and is not required by the default flow
- while downloading, new messages receive a wait prompt, and the state resets to IDLE after completion
It does not accept full cookies or create config/cookies.config. Login state is determined by the host environment’s BiliDownloader: without login, it typically requests up to 720p video and high-resolution audio; with login, it can request up to 8K video, with final quality depending on source, account permissions, and API responses. Trigger phrases include 下载B站视频, 下载B站音频, 保存BV, and bilibili下载. Audio tasks prefer .m4a and fall back to .mp3 if extraction is unavailable; video tasks default to .mp4. Do not run concurrent downloads in the same skill instance, and do not leave stale pending state after completion, cancellation, or errors.
Use Cases
- Extract a Bilibili BV ID from chat and let the agent call the download interface to save the video locally.
- When asked to save Bilibili audio, output .m4a first and fall back to .mp3 if extraction is unavailable.
- Download multi-part videos by passing all pages or selected pages through the pages parameter instead of manual confirmation.
- Use a host-provided logged-in BiliDownloader to request higher quality and handle network or FFmpeg errors.
Best For
- Python engineers maintaining BiliDownloader who want to wrap BV parsing, multi-P downloads, and error messages as agent-callable interfaces.
- Product engineers building conversational tools who need trigger-based Bilibili video or audio downloads with stable state feedback.
- Local media automation authors who need wait prompts during download, state reset after completion, and clear 720p/8K login boundaries.
- Agent skill integrators who want to expose only media_type and pages, without accepting cookies or creating local config files.
Related Skills
Turn vague requests into structured prompts with role, task, context, and constraints, outputting only usable prompts without executing the task.
A skill for game NPC and enemy AI that separates decision, steering, and pathfinding, covering FSMs, behavior trees, A*, path caching, and pitfalls.
Provides multi-turn emotional companion chat through a third-party Agent API and supports continuing sessions via session_id.
An evaluation-driven workflow for diagnosing, optimizing, validating, and logging Agent Skills based on design patterns and anti-patterns.