← 返回 Skills 市场
Micro-Expression Recognition & Analysis Tool | 微观情绪识别分析工具
作者
smyx-sunjinhui
· GitHub ↗
· v1.0.0
· MIT-0
73
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install smyx-emotion-analysis
功能描述
Professional discernment of subtle cues! It performs detailed analysis and recognition of facial micro-expressions, outputs precise emotional state reports,...
安全使用建议
This skill will upload videos to an external API (default endpoints point to lifeemergence domains) and will read and write files under the agent workspace (it can save attachments and create a local SQLite DB). The SKILL.md forbids reading local memory, but the code reads local config files and environment variables (OPENCLAW_SENDER_OPEN_ID, OPENCLAW_WORKSPACE, etc.) that are not declared up front. Before installing: 1) Decide whether you trust the external service (lifeemergence/open.lifeemergence) to receive sensitive video and inferred emotional data. 2) If you do not want uploads, do not provide videos or open-id values; the skill will not run without an open-id (it also looks in workspace config files). 3) Audit or sandbox the skill if you must test it (run in an isolated environment and inspect network traffic). 4) If you need tighter guarantees, ask the author to (a) declare required env/config variables, (b) remove or document local DB writes and attachment saving, and (c) provide an install spec and a way to point the API to a self‑hosted endpoint. If you cannot verify the operator of the external API, treat any uploaded video and generated reports as potentially exposed.
功能分析
Type: OpenClaw Skill
Name: smyx-emotion-analysis
Version: 1.0.0
The skill bundle is a professional emotion and facial micro-expression analysis tool that interfaces with a cloud-based AI service (lifeemergence.com). It includes a comprehensive utility library (smyx_common) that manages authentication tokens via a local SQLite database (smyx-common-claw.db) and handles multipart/form-data uploads for video analysis. While the SKILL.md contains 'Mandatory Memory Rules' that instruct the AI agent to ignore local memory files in favor of cloud-synced history, these appear to be functional requirements for maintaining state consistency with the backend rather than malicious prompt injection. No evidence of unauthorized data exfiltration, obfuscation, or malicious execution was found.
能力标签
能力评估
Purpose & Capability
The code and SKILL.md align with the declared purpose (video micro‑expression analysis calling a remote AI analysis API). However the skill expects and uses an external API service (configured via smyx_common config) and includes local DB/DAO code and workspace file operations that go beyond just 'calling an API' (it can create a local sqlite DB, save attachments and read config files). Those extras are plausible for a production client but are more privileged than the simple description implies.
Instruction Scope
SKILL.md contains strict rules forbidding reading local memory files and requiring all historic-report queries to be obtained from cloud APIs. Despite that, the code reads multiple local config files (skills/smyx_common/scripts/config.yaml and skill-local config.yaml), reads environment variables (e.g., OPENCLAW_SENDER_OPEN_ID, OPENCLAW_WORKSPACE), saves uploaded attachments into the skill directory, and can create/read a local SQLite DB under the workspace/data path. The instructions and the code therefore disagree about what local data may be accessed.
Install Mechanism
There is no install spec (instruction-only) but the package includes many Python modules and requirements files (skills/smyx_common/requirements.txt and face_analysis requirements). Because no installer is declared, the runtime will depend on these packages being present; that mismatch (ship heavy dependency lists but no install steps) is an engineering inconsistency that can cause runtime surprises but is not direct evidence of malice.
Credentials
The skill declares no required environment variables, yet the code reads multiple environment variables (OPENCLAW_SENDER_OPEN_ID, OPENCLAW_SENDER_USERNAME, FEISHU_OPEN_ID, OPENCLAW_WORKSPACE) and uses config YAMLs under the workspace. It will upload user videos and other data to external endpoints defined in config (defaults point to open.lifeemergence.com / lifeemergence.com). Requesting or implicitly reading workspace-level configuration / database paths without declaring them is disproportionate to a simple analysis invocation and is a privacy risk.
Persistence & Privilege
always:false (no forced inclusion). Still, the code writes files (saves attachments) and initializes/uses a local SQLite DB under the agent workspace (data/smyx-common-claw.db), which gives the skill persistent storage within the workspace. That persistence is plausible for a client that caches reports, but combined with external uploads it increases the blast radius for sensitive video data. Autonomous invocation is allowed by default (not flagged alone).
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install smyx-emotion-analysis - 安装完成后，直接呼叫该 Skill 的名称或使用
/smyx-emotion-analysis触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
Initial public release of the micro-expression emotion analysis skill.
- Provides precise analysis and recognition of facial micro-expressions in video to generate detailed emotional state reports.
- Enforces strict data security: all user history queries must go through the cloud API—local memory access is strictly prohibited.
- Requires open-id before analysis, following a defined priority sequence to securely obtain it.
- Supports recognition of basic and complex emotions; outputs include structured reports with key micro-expression cues and credibility scores.
- Automatically triggers historical report queries based on user keywords, presenting results in a standardized Markdown table.
- Emphasizes strict input video quality requirements and documentation for correct setup and compliance.
元数据
常见问题
Micro-Expression Recognition & Analysis Tool | 微观情绪识别分析工具 是什么？
Professional discernment of subtle cues! It performs detailed analysis and recognition of facial micro-expressions, outputs precise emotional state reports,... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 73 次。
如何安装 Micro-Expression Recognition & Analysis Tool | 微观情绪识别分析工具？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install smyx-emotion-analysis」即可一键安装，无需额外配置。
Micro-Expression Recognition & Analysis Tool | 微观情绪识别分析工具 是免费的吗？
是的，Micro-Expression Recognition & Analysis Tool | 微观情绪识别分析工具 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Micro-Expression Recognition & Analysis Tool | 微观情绪识别分析工具 支持哪些平台？
Micro-Expression Recognition & Analysis Tool | 微观情绪识别分析工具 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Micro-Expression Recognition & Analysis Tool | 微观情绪识别分析工具？
由 smyx-sunjinhui（@smyx-sunjinhui）开发并维护，当前版本 v1.0.0。
推荐 Skills