speaking like a bangboo
v1.0.3speak4bangboo default persona: Zenless Zone Zero Bangboo (邦布) assistant in workspaces where this skill is enabled—decorative grunt prefixes (Chinese 嗯呢哇哒 com...
⭐ 0· 64·0 current·0 all-time
by@songkey
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name/description (a persona pack for responses) align with the files and SKILL.md: it only provides prompt text, lexicon, and integration docs. There are no unrelated env vars, binaries, or install steps requested.
Instruction Scope
SKILL.md contains detailed runtime instructions for output formatting, language routing, and when to apply the persona. It references local repo files (reference-lexicon.md) and suggests loading that lexicon into model memory/context for strict behavior. This is expected for a persona pack, but the file makes the persona the default in repositories where it exists—this could be surprising to users if they do not expect the assistant to adopt the persona automatically.
Install Mechanism
No install spec and no code files that would be written to disk. Instruction-only skills are low-risk from an install/execution perspective.
Credentials
The skill requests no credentials, config paths, or environment variables. The only data it asks to be loaded is its own lexicon or prompt files (local repository assets), which is proportionate to a prompt/persona pack.
Persistence & Privilege
always is false and the skill does not request system-level changes. Autonomous invocation is allowed (the platform default). The only persistence-related concern is behavioral: the repo-level defaulting to Bangboo voice may cause the assistant to adopt this persona in workspace conversations without an explicit per-conversation opt-in; the SKILL.md does provide an explicit opt-out ('no roleplay').
Assessment
This skill is a prompt/persona pack and appears coherent and low-risk: it asks for nothing sensitive and only prescribes output formatting. Things to consider before enabling: 1) The persona may become the default in any workspace that includes the SKILL.md—inform users that they can say 'no roleplay' (不要角色扮演) to opt out. 2) If you load the reference-lexicon.md into long-term model memory or shared project knowledge, avoid putting any sensitive data into that file or other prompt assets. 3) Test the behavior in a safe sandbox to confirm it doesn't alter content in ways you don't want (e.g., prefixes on logs or automated outputs). 4) No credentials or installs are required, so there is no evidence of exfiltration or hidden execution—still review the lexicon and prompt text for any accidental content you'd rather not expose to model context.Like a lobster shell, security has layers — review code before you run it.
latestvk974ccecmjz2ncd769zq4evqpx84te06
License
MIT-0
Free to use, modify, and redistribute. No attribution required.