ClawHub

Bilibili Messager | B站私信助手

v1.3.5

Send Bilibili DMs, reply, and read chat history through browser automation. 自动发送 B 站私信、回复消息、读取聊天记录。

⭐ 0· 690·3 current·3 all-time
by@moroiser
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name/description (send/read Bilibili DMs via browser automation) align with the content of SKILL.md: all guidance is about opening the Bilibili messages page, DOM inspection, and sending after explicit confirmation. No unrelated credentials, binaries, or installs are requested.
✓
Instruction Scope
Instructions only reference browser actions, DOM queries, snapshots, and verification steps. The SKILL.md explicitly requires the user to be logged in in the browser profile and insists on user confirmation before sending. Note: snapshots capture page content (potentially sensitive chat data) — the doc already advises sparing use, but users should be aware of that risk.
✓
Install Mechanism
No install spec and no code files — instruction-only. This minimizes disk/write risk; nothing is downloaded or installed.
✓
Credentials
No environment variables, credentials, or config paths are requested. The skill relies on the user's existing logged-in browser session, which is proportionate to the stated function.
✓
Persistence & Privilege
always is not set and the skill does not request elevated/system-wide persistence. Model invocation is allowed (default), which is normal — be mindful that an autonomous agent with this skill could attempt actions in the browser if granted permission by the user at runtime.
Assessment
This skill appears coherent for automating Bilibili private messages via a logged-in browser profile and requests no extra credentials or installs. Before installing: (1) confirm you are comfortable with the agent accessing your browser/profile session (it will read DOM and may take snapshots), (2) ensure the agent is configured to require explicit user confirmation before sending messages, and (3) monitor snapshot usage because snapshots can include private chat content. If you want stronger safeguards, keep autonomous invocation disabled or require manual approval for actions that send messages.

Like a lobster shell, security has layers — review code before you run it.

latestvk97114kdz93d64x8ga5qsbk54184a5hv

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Comments