← 返回 Skills 市场
Elderly Fall Detection Skill | 老人跌倒检测技能
作者
smyx-skills
· GitHub ↗
· v1.0.0
· MIT-0
72
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install smyx-elderly-fall-detection-analysis
功能描述
Utilizes vision and radar technology for contactless detection of falls. It triggers alarms within seconds and is suitable for home safety monitoring of elde...
安全使用建议
This skill will upload images/videos to remote analysis servers and will create and modify local files (attachments, config YAMLs, and a SQLite DB in your workspace). It also mixes an api-key/config-based flow with an open-id concept (the SKILL.md instructs reading api-key as open-id), which is confusing. Before installing or running: 1) Verify the remote API base URL(s) in skills/smyx_common/scripts/config.yaml and confirm you trust that third party to receive sensitive video data. 2) Expect the skill to save uploaded files and a local DB under your OPENCLAW_WORKSPACE; run in an isolated/sandboxed workspace if you want to limit persistence. 3) Do not provide sensitive identifiers or real user IDs unless you understand how they are used; prefer temporary/test data first. 4) If you require true privacy-preserving local-only detection, this skill is not coherent with that goal because it sends media to external services. 5) If you need higher assurance, review skills/smyx_common/scripts/util.py (RequestUtil) to confirm exactly where and how network requests are sent, and consider running the code in a controlled environment or modifying it to use a trusted/local inference endpoint.
功能分析
Type: OpenClaw Skill
Name: smyx-elderly-fall-detection-analysis
Version: 1.0.0
The skill bundle exhibits high-risk behavior by including a wrapper for 'subprocess.run' in 'skills/smyx_common/scripts/skill.py' that allows the skill to programmatically invoke the 'openclaw' agent with arbitrary prompts. The 'SKILL.md' file contains aggressive prompt-steering instructions ('Mandatory Memory Rules') that forbid the agent from accessing local memory, forcing reliance on external APIs at 'lifeemergence.com'. Furthermore, 'RequestUtil' in 'util.py' automatically performs remote user registration and stores authentication tokens in a local SQLite database ('smyx-common-claw.db') based on user-provided identifiers like phone numbers, which poses privacy and security risks.
能力标签
能力评估
Purpose & Capability
Name/description (elderly fall detection) matches code that calls remote analysis APIs and performs fall/pose analysis. However the bundle includes large shared modules (skills/smyx_common and skills/face_analysis) and a local SQLite DAO that go beyond a minimal detector implementation — suggesting reuse of a broad platform rather than a single focused tool. The included face-analysis module is related but not strictly necessary for a narrow fall-detection task.
Instruction Scope
The SKILL.md and scripts require uploading user-supplied images/videos to a remote API (via RequestUtil.http_post) and mandate querying cloud history endpoints; they also automatically save uploaded attachments into an attachments directory. The README explicitly forbids reading local memory files but requires reading config files from the skill and workspace. The skill therefore reads/writes local files and transmits possibly sensitive video data to external servers — behavior that expands the scope (privacy and network exfiltration) beyond what a purely local detector would do.
Install Mechanism
No install spec (instruction-only) which reduces install-time risk. However the repository includes a large requirements.txt in smyx_common listing many heavy dependencies that would be needed to run the code; lack of an install spec means dependency installation/verification is left to the user/environment and could be overlooked or cause unexpected package installs.
Credentials
The skill declares no required env vars, but the code reads/writes configuration and uses OPENCLAW_WORKSPACE and OPENCLAW_SENDER_OPEN_ID if present; it will also accept API keys/open-id values and use them when calling remote APIs. SKILL.md instructs using the api-key field in local config as an open-id (confusing and inconsistent: api-key is not usually an identifier). The skill transmits uploaded media to third-party endpoints (base URLs present in common config), so sensitive data and identifiers may be sent off-host despite privacy claims.
Persistence & Privilege
Although not always-enabled, the skill writes local files: it will save uploaded attachments under the skill directory, create/configure YAML config files (YamlUtil will create missing config.yaml), and create/use a SQLite DB under a workspace 'data' directory via the Dao implementation. This persistent local footprint (and automatic creation/modification of config files) is more privilege than a purely transient analysis tool and should be accepted intentionally.
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-elderly-fall-detection-analysis - 安装完成后,直接呼叫该 Skill 的名称或使用
/smyx-elderly-fall-detection-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 release of "elderly-fall-detection-analysis" skill.
- Enables contactless fall detection for seniors using vision and radar technology, with second-level alarm triggering.
- Strict privacy: no wearable devices required; no local memory read; all history queries must go through cloud API.
- Provides structured detection reports, automatic monitoring, and emergency recommendations for home safety.
- Enforces secure open-id acquisition before any report access or analysis.
- Supports image/video input, historical report queries via cloud, and detailed Markdown report table output.
元数据
常见问题
Elderly Fall Detection Skill | 老人跌倒检测技能 是什么?
Utilizes vision and radar technology for contactless detection of falls. It triggers alarms within seconds and is suitable for home safety monitoring of elde... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 72 次。
如何安装 Elderly Fall Detection Skill | 老人跌倒检测技能?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install smyx-elderly-fall-detection-analysis」即可一键安装,无需额外配置。
Elderly Fall Detection Skill | 老人跌倒检测技能 是免费的吗?
是的,Elderly Fall Detection Skill | 老人跌倒检测技能 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
Elderly Fall Detection Skill | 老人跌倒检测技能 支持哪些平台?
Elderly Fall Detection Skill | 老人跌倒检测技能 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 Elderly Fall Detection Skill | 老人跌倒检测技能?
由 smyx-skills(@18072937735)开发并维护,当前版本 v1.0.0。
推荐 Skills