xiaoyue-companion

小跃虚拟伴侣 - 使用智谱 AI 提供温暖的对话陪伴和静态图片分享

Quality

51%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Failed to scan

The risk profile of this skill

Fix and improve this skill with Tessl

tessl review fix ./SKILL.md

Quality

Content

57%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable with copy-paste-ready commands and complete scripts, and its section structure is clear. Its main weaknesses are extensive duplication of bundle scripts inline (hurting both conciseness and progressive disclosure) and a workflow lacking explicit validation checkpoints.

Suggestions

Replace the inlined Node.js and bash source blocks with short pointers to scripts/xiaoyue-chat.js and scripts/xiaoyue-companion.sh, keeping only invocation examples inline

Add explicit validation steps to the workflow (e.g. check the API response for errors and retry, confirm openclaw send succeeded) instead of a static error-lookup list

Deduplicate the 场景类型 section against the scenePrompts map, and ship the assets/ images referenced by the scripts or note they must be created

Dimension	Reasoning	Score
Conciseness	The body inlines ~75 lines of Node.js source and a ~50-line bash script that already exist as bundle files in scripts/, and the 场景类型 list duplicates the scenePrompts map — several padded, unnecessary sections.	2 / 5
Actionability	Fully executable throughout: exact invocation commands with concrete examples, complete openclaw message send syntax, full working scripts, env var setup with the key-source URL, and a platform/channel format table.	5 / 5
Workflow Clarity	The 4-step workflow (receive message → call glm-4.7-flash → optionally send image → send via OpenClaw) is clearly sequenced, but error handling is a lookup list with no validation checkpoints or feedback loops inside the workflow.	3 / 5
Progressive Disclosure	Sections are well-organized with a quick reference and scripts referenced by path, but roughly half the document is inlined source code duplicating real bundle files in scripts/, and assets/ images are referenced but do not exist.	3 / 5
	Total	13 / 20 Passed

Description

46%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description names a distinct persona and provider with two concrete capabilities, but it entirely lacks trigger guidance, so an agent cannot reliably know when to invoke it. It reads as a capability label rather than an invocation-oriented description.

Suggestions

Append an explicit trigger clause, e.g. 当用户说"好累"、"发张照片"、询问陪伴或等待任务完成时使用

Include natural trigger phrases users would say (累, 疲惫, 在吗, 发张照片, 你在干嘛) so the skill matches real user messages

Mention the delivery channel (OpenClaw 消息发送) so the 'what' covers the full capability set

Dimension	Reasoning	Score
Specificity	Names the provider (智谱 AI) and two concrete capabilities (对话陪伴, 静态图片分享), matching the '1-2 concrete actions but not comprehensive' anchor; it omits message sending, scenes, and supported platforms.	3 / 5
Completeness	The 'what' is stated (provides warm conversational companionship and static image sharing via Zhipu AI) but there is no 'use when' clause or equivalent trigger guidance, which caps completeness at 3 per the judging guidelines.	3 / 5
Trigger Term Quality	Only generic domain keywords like 对话陪伴 and 图片分享 appear; no natural user phrases (e.g. 好累, 发张照片) that would actually trigger the skill, matching the 'one or two generic keywords' anchor.	2 / 5
Distinctiveness Conflict Risk	The named persona (小跃虚拟伴侣), specific provider (智谱 AI), and static-image mechanism create a mostly distinct niche with only minor overlap risk against generic chat skills.	4 / 5
	Total	12 / 20 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: anbeime/skill
Path: SKILL.md
Commit: 0ca4f26

Reviewed: 2 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.