← 返回 Skills 市场
Dietary Behavior Health Analyzer | 饮食行为健康分析工具
作者
smyx-sunjinhui
· GitHub ↗
· v1.0.0
· MIT-0
66
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install smyx-diet-analysis
功能描述
Analyzes videos to evaluate human eating behaviors, habits, and dietary patterns. It identifies tendencies towards unhealthy eating and provides structured a...
安全使用建议
Key points to confirm before installing or running this skill:
- The skill will upload user videos to external API endpoints (API base URLs are present in config YAMLs). Confirm you trust the remote service and its domain (lifeemergence/open APIs shown in configs).
- Although the SKILL.md says no local memory usage, the code contains a DAO and will create/read config.yaml and an SQLite DB under the workspace/data path; ask the maintainer whether any user data or reports are stored locally and for how long.
- The skill expects workspace-related environment variables (e.g. OPENCLAW_WORKSPACE) and may read OPENCLAW_SENDER_OPEN_ID / FEISHU_OPEN_ID to set the current open-id, but the registry metadata declares no required env vars. Provide or restrict those env vars consciously and verify the open-id acquisition flow.
- The package includes a large requirements list but has no install spec—decide whether to run in a sandboxed environment (container) and inspect runtime network traffic (or run a dry-run) to confirm endpoints and what is transmitted.
- Request clarification from the publisher on: (1) where and for how long user videos/reports are stored (local DB vs remote), (2) exact external endpoints used and data retention policy, and (3) an explicit list of required environment variables and permissions. If you cannot get clear answers, run the skill in an isolated environment or avoid installing it.
功能分析
Type: OpenClaw Skill
Name: smyx-diet-analysis
Version: 1.0.0
The skill bundle contains high-risk capabilities and unusual structural patterns. Specifically, 'skills/smyx_common/scripts/skill.py' includes logic to execute the 'openclaw agent' CLI via 'subprocess.run', which allows the skill to trigger broader agent actions. The 'SKILL.md' file uses high-priority 'Mandatory Rules' to force the AI agent to ignore local memory/LanceDB and exclusively use specific external API calls (targeting lifeemergence.com), which is a form of prompt control that redirects data flow. Furthermore, the bundle is significantly bloated, including an entirely separate 'face_analysis' skill and a complex common library that manages local SQLite databases for token persistence, which exceeds the requirements for a simple diet analysis tool.
能力标签
能力评估
Purpose & Capability
The skill claims no required environment variables or binaries, yet the codebase reads environment variables (e.g. OPENCLAW_WORKSPACE, OPENCLAW_SENDER_OPEN_ID, FEISHU_OPEN_ID) and relies on common modules that manage API endpoints and a local SQLite DB. Uploading user videos to an external API is coherent with the stated purpose, but the presence of a local DAO, database file paths, and utilities that create config files suggests additional local persistence not described in the SKILL.md.
Instruction Scope
SKILL.md explicitly forbids reading local 'memory' and LanceDB and mandates fetching historical reports from cloud APIs only; however the included code contains a local DAO/SQLite implementation, utilities that create/read config.yaml files, and logic that sets/reads CURRENT__OPEN_ID from environment or config. The instructions also require reading config files under ${OPENCLAW_WORKSPACE}, but the skill metadata did not declare that env var is required—this is an inconsistency between documented runtime rules and actual code behavior.
Install Mechanism
This is instruction-plus-code (no install spec). The package includes a large common requirements list (skills/smyx_common/requirements.txt) with many third-party packages, but there is no install specification or sandboxing. That increases friction and risk (missing dependency installation instructions, unexpected transitive packages), but there is no direct evidence of a malicious install mechanism (no remote download URLs in the manifest).
Credentials
Declared requirements list no env vars, but code depends on environment variables and config files (OPENCLAW_WORKSPACE, OPENCLAW_SENDER_OPEN_ID, OPENCLAW_SENDER_USERNAME, FEISHU_OPEN_ID, ApiEnum.BASE_URL_* from config YAMLs). The skill also expects an open-id (user identifier) and may accept API keys; these are reasonable for a cloud API integration, but the omission of those env variables from the metadata is a mismatch that could lead to surprising behavior or data placement.
Persistence & Privilege
Although always:false and model invocation is allowed by default, the code will create/read config.yaml files and initialize/use a local SQLite DB under a workspace 'data' directory (skills/smyx_common.scripts.dao.get_db_path). SKILL.md forbids local memory usage for report queries, yet the codebase clearly contains local persistence utilities that may store records and create files—this grants the skill more persistent local presence than documented.
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-diet-analysis - 安装完成后，直接呼叫该 Skill 的名称或使用
/smyx-diet-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
diet-analysis v1.0.0 initial release:
- Provides automated video analysis of eating behaviors, habits, and dietary patterns.
- Identifies unhealthy eating tendencies and generates structured analysis reports with nutrition recommendations.
- Enforces strict rules to only use cloud interfaces for report history (never read local memory or fallback).
- Requires open-id (api-key, username, or phone number) for all operations, following a specific retrieval sequence.
- Supports comprehensive and specialized analyses (speed, habit, structure, risk) via simple command-line options.
- Outputs history results as interactive Markdown tables with direct report links.
元数据
常见问题
Dietary Behavior Health Analyzer | 饮食行为健康分析工具 是什么？
Analyzes videos to evaluate human eating behaviors, habits, and dietary patterns. It identifies tendencies towards unhealthy eating and provides structured a... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 66 次。
如何安装 Dietary Behavior Health Analyzer | 饮食行为健康分析工具？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install smyx-diet-analysis」即可一键安装，无需额外配置。
Dietary Behavior Health Analyzer | 饮食行为健康分析工具 是免费的吗？
是的，Dietary Behavior Health Analyzer | 饮食行为健康分析工具 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Dietary Behavior Health Analyzer | 饮食行为健康分析工具 支持哪些平台？
Dietary Behavior Health Analyzer | 饮食行为健康分析工具 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 Dietary Behavior Health Analyzer | 饮食行为健康分析工具？
由 smyx-sunjinhui（@smyx-sunjinhui）开发并维护，当前版本 v1.0.0。
推荐 Skills