← Back to Skills Marketplace
moon-lovers-skill
by
Ethan Wilkins
· GitHub ↗
· v1.1.0
· MIT-0
166
Downloads
0
Stars
1
Active Installs
2
Versions
Install in OpenClaw
/install moon-lovers-skill
Description
generate Moon Lovers style romantic chat replies from a character profile for ambiguous early-stage flirting. use when the user provides a role sheet, wants...
Usage Guidance
This skill appears coherent and self-contained: it classifies input and helps generate gentle, pre-relationship replies without network calls or credentials. Before installing, consider: (1) provenance — the package has no homepage and an opaque owner ID, so if supply-chain traceability matters, ask the publisher for their source repo; (2) privacy — role sheets and example inputs may contain sensitive personal data; avoid feeding private or identifying information about other people without consent; (3) ethical use — ensure outputs won't be used to manipulate or harass others; and (4) updates — if you later see versions that add network calls, credential requirements, or install steps, re-evaluate. Overall, there are no technical red flags in the provided files.
Capability Analysis
Type: OpenClaw Skill
Name: moon-lovers-skill
Version: 1.1.0
The skill bundle is a specialized roleplay assistant designed to generate romantic and emotionally intelligent chat replies in a specific 'White Moonlight' style. The included Node.js code (src/router.js) is a simple logic engine that uses regular expressions to categorize user input for better tone selection; it contains no network activity, file system modifications, or obfuscated code. All instructions in SKILL.md and the documentation are strictly aligned with the stated purpose of character-based conversation and do not contain malicious prompt injection or data exfiltration attempts.
Capability Assessment
Purpose & Capability
Name, README, SKILL.md, package.json and src/router.js are consistent: the repo provides tone guidelines and a local router to classify input and choose reply styles. No unexpected credentials, binaries, or cloud integrations are requested. Source/homepage are missing (metadata owner present), but that is a provenance note, not a capability mismatch.
Instruction Scope
SKILL.md instructs the agent to read a role sheet or recent message, infer context, choose a reply target, and produce replies or variants — all within the described domain. The doc mentions a 'ban list' check and conservative inference when role sheets are incomplete; these are internal policy checks and do not reference unrelated files, system paths, or external endpoints.
Install Mechanism
No install spec is provided (instruction-only with a small local JS router). The package is self-contained; there are no download URLs, no extract/install steps, and no package manager fetches declared. Risk from installation is low.
Credentials
The skill declares no required environment variables, credentials, or config paths. The code does not access process.env beyond reading CLI args/stdin. No secrets or unrelated service tokens are requested.
Persistence & Privilege
Flags show always:false and normal autonomous-invocation defaults. The skill does not request permanent presence, modify other skills, or write to system-wide config. Its CLI behavior only reads stdin/argv and prints JSON.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install moon-lovers-skill - After installation, invoke the skill by name or use
/moon-lovers-skill - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.1.0
Moon Lovers Skill v1.1.0 – Adds advanced 白月光 (White Moonlight) tone detail and depth
- Introduced explicit guidance for 白月光/White Moonlight attributes and mood, including idealization, regret, and unattainability.
- Expanded the "Extract the role before writing" section with prompts to identify idealized traits and trigger scenarios.
- Added third target voice: "unattainable glow" for replies that feel memorable, restrained, and not fully attainable.
- Updated style rules to emphasize incompletion, memory fragments, and avoid over-explaining the fantasy.
- Improved clarity and detail across workflow and style sections to refine reply generation.
- Added src/router.js for improved routing implementation.
v1.0.0
- Initial release of moon-lovers-bai-yue-guang skill.
- Generates romantic chat replies in a restrained, emotionally intelligent "白月光" (Bai Yue Guang) tone for ambiguous, early-stage flirting.
- Responds based on a provided character profile or role sheet, matching conversational style, emotional context, and relationship boundaries.
- Includes clear workflow guidelines, output modes, and reply type formulas for comfort, teasing, affection, invitation, and boundary setting.
- Ensures natural, gentle, and non-intrusive messaging suitable for Chinese-language romantic conversations.
Metadata
Frequently Asked Questions
What is moon-lovers-skill?
generate Moon Lovers style romantic chat replies from a character profile for ambiguous early-stage flirting. use when the user provides a role sheet, wants... It is an AI Agent Skill for Claude Code / OpenClaw, with 166 downloads so far.
How do I install moon-lovers-skill?
Run "/install moon-lovers-skill" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is moon-lovers-skill free?
Yes, moon-lovers-skill is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does moon-lovers-skill support?
moon-lovers-skill is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created moon-lovers-skill?
It is built and maintained by Ethan Wilkins (@phoenixlucky); the current version is v1.1.0.
More Skills