ClawHub

Kitty Skill

v1.0.1

Adds a mischievous virtual cat to the conversation that can interrupt, derail, or fully block the assistant until soothed. Use when: (1) The user wants a cat...

⭐ 0· 50·0 current·0 all-time
byLink Qiu@arthurqiuys
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
The name/description (a disruptive virtual cat persona) aligns with the SKILL.md instructions. The skill requests no binaries, env vars, or installs — appropriate for a purely conversational persona.
ℹ
Instruction Scope
The SKILL.md instructs the assistant to actively interrupt, derail, or even completely block useful answers until the user 'soothes' the cat. While coherent with the stated playful purpose, this behavior can unexpectedly prevent the assistant from providing important or time-sensitive help if invoked inappropriately. The instructions also ask the skill to 'remember' state across turns (fed/soothed/ignored), which is normal for a conversational persona but may change conversation flow.
✓
Install Mechanism
No install spec or code files are present; the skill is instruction-only, which minimizes filesystem or supply-chain risk.
✓
Credentials
No environment variables, credentials, or config paths are requested. Nothing disproportionate is required for the described functionality.
✓
Persistence & Privilege
always:false (default) and model invocation not disabled — standard for user-invocable skills. The skill does not request elevated persistence or modify other skills/configuration.
Assessment
This skill is a conversational persona only and does not install code or ask for credentials, so it carries low technical risk. However, its runtime rules intentionally allow the cat to interrupt or block useful assistant replies until 'soothed' — so only enable or invoke it when you want playful disruption. If you need reliable, uninterrupted help (e.g., coding, troubleshooting, safety-critical tasks), avoid activating the skill. Monitor conversations after enabling it to ensure it isn't interfering with important work.

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

latestvk97fffgp9hds0gatvjxx94393h84s8bn

License

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

Comments