← Back to Skills Marketplace
TencentCloud MLIDPassport OCR
by
tencent-ocr
· GitHub ↗
· v1.0.2
· MIT-0
383
Downloads
1
Stars
1
Active Installs
3
Versions
Install in OpenClaw
/install tencentcloud-ocr-mlidpassport
Description
腾讯云护照识别（多国多地区）(MLIDPassportOCR)接口调用技能。当用户需要识别护照图片中中国大陆、港澳台地区或其他国家/地区的护照信息（护照ID、姓名、出生日期、性别、有效期、发行国、国籍、国家地区代码、MRZ码等）时,应使用此技能。支持图片Base64和URL两种输入方式,支持护照图片人像照片裁剪功...
Usage Guidance
This skill implements Tencent Cloud passport OCR and needs your Tencent Cloud API keys and will send image data to ocr.tencentcloudapi.com. Before installing: (1) Be aware the tool accepts a file path for --image-base64 and will read and transmit any file you point it at — avoid passing sensitive local file paths. (2) Provide a least‑privilege Tencent Cloud account (monitor and restrict API permissions, use billing/usage limits). (3) Verify and correct metadata: the registry should declare the required environment variables (TENCENTCLOUD_SECRET_ID and TENCENTCLOUD_SECRET_KEY). (4) If you intend to let an autonomous agent call this skill, add explicit guardrails (don’t allow arbitrary path inputs, require user confirmation, or whitelist input directories). (5) If you have a data residency or privacy requirement, note that images (and any file contents) will be sent to Tencent Cloud endpoints and possibly to regions you select via --region.
Capability Analysis
Type: OpenClaw Skill
Name: tencentcloud-ocr-mlidpassport
Version: 1.0.2
The skill bundle provides a standard integration with Tencent Cloud's Passport OCR service using the official Python SDK. It correctly handles authentication via environment variables (TENCENTCLOUD_SECRET_ID/KEY) and communicates only with the legitimate Tencent Cloud endpoint (ocr.tencentcloudapi.com). No evidence of malicious intent, data exfiltration, or prompt injection was found in scripts/main.py or SKILL.md.
Capability Assessment
Purpose & Capability
The name, description, SKILL.md and the included script all implement Tencent Cloud MLIDPassportOCR functionality and require Tencent Cloud API keys — that capability is coherent with the purpose. However the registry metadata at the top claims 'Required env vars: none' and 'Primary credential: none' while the SKILL.md and script explicitly require TENCENTCLOUD_SECRET_ID and TENCENTCLOUD_SECRET_KEY. This metadata omission is an inconsistency that should be corrected before trusting the package.
Instruction Scope
The runtime instructions tell the agent to run scripts/main.py which accepts --image-base64 either as a Base64 string or as a file path; the script will read any provided local file, (base64-)encode it and send it to Tencent Cloud. That behavior is reasonable for image files, but it also enables sending arbitrary local file contents to the external OCR endpoint if a path to a non-image file is provided. The SKILL.md does not explicitly restrict inputs to safe/expected locations or warn about this exfiltration vector. Additionally, the script will transmit data to ocr.tencentcloudapi.com, including any provided image content and the configured credentials' usage.
Install Mechanism
This is an instruction-only skill with no install spec (lowest install risk). It depends on the public tencentcloud-sdk-python package but does not download arbitrary code or write files during install.
Credentials
Requesting Tencent Cloud API keys (TENCENTCLOUD_SECRET_ID and TENCENTCLOUD_SECRET_KEY) is proportionate to calling the Tencent OCR API. However the package metadata fails to declare these required environment variables (metadata shows none), which is misleading. The script does not request additional unrelated credentials.
Persistence & Privilege
The skill does not request always:true and does not modify other skills or system-wide configuration. It runs on-demand and has normal autonomous invocation settings (disable-model-invocation=false), which is the platform default.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install tencentcloud-ocr-mlidpassport - After installation, invoke the skill by name or use
/tencentcloud-ocr-mlidpassport - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.2
- 新增 UserAgent（可选）参数，用于请求来源标识，统一为 Skills
- 其余内容保持不变，无其他功能或接口变动
v1.0.1
更新显示名称
v1.0.0
tencentcloud-ocr-mlidpassport v1.0.0
- Initial release of the skill for Tencent Cloud OCR multi-country passport recognition.
- Supports extraction of key passport information from images or URLs, including name, passport number, nationality, MRZ code, and more.
- Handles passport photo cropping (optional) and supports over 80 countries/regions.
- Includes international site-exclusive alarms for image tampering, photocopies, reflections, blurriness, and incomplete borders.
- Provides error and warning code documentation, usage scenarios, and example command-line calls.
Metadata
Frequently Asked Questions
What is TencentCloud MLIDPassport OCR?
腾讯云护照识别（多国多地区）(MLIDPassportOCR)接口调用技能。当用户需要识别护照图片中中国大陆、港澳台地区或其他国家/地区的护照信息（护照ID、姓名、出生日期、性别、有效期、发行国、国籍、国家地区代码、MRZ码等）时,应使用此技能。支持图片Base64和URL两种输入方式,支持护照图片人像照片裁剪功... It is an AI Agent Skill for Claude Code / OpenClaw, with 383 downloads so far.
How do I install TencentCloud MLIDPassport OCR?
Run "/install tencentcloud-ocr-mlidpassport" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is TencentCloud MLIDPassport OCR free?
Yes, TencentCloud MLIDPassport OCR is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does TencentCloud MLIDPassport OCR support?
TencentCloud MLIDPassport OCR is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created TencentCloud MLIDPassport OCR?
It is built and maintained by tencent-ocr (@zt1314p-design); the current version is v1.0.2.
More Skills