← Back to Skills Marketplace
Scnet Ocr
by
SCNet-sugon
· GitHub ↗
· v1.0.5
· MIT-0
262
Downloads
0
Stars
0
Active Installs
4
Versions
Install in OpenClaw
/install scnet-ocr
Description
将图片中的文字、通用文字识别, 居民身份证, 银行卡, 社保卡, 户口本, 出生医学证明, 往来港澳通行证, 往来台湾通行证, 台湾居民来往大陆通行证, 港澳居民来往内地通行证, 营业执照, 社会团体法人登记证书, 工会法人资格证书, 宗教活动场所登记证, 民办非企业单位登记证书, 事业单位法人证书, 统一社会信...
Usage Guidance
This skill appears to be a straightforward client for the Scnet OCR API. Before installing, consider: 1) Keep your SCNET_API_KEY secret — do not paste it in chat; store it in the recommended config/.env with file mode 600 or set it as an environment variable. 2) When invoking the skill, only provide image/PDF file paths you intend to upload; the script will read and send any file path you give to the remote API (so do not point it at sensitive system files). 3) Confirm the SCNET API endpoint (default https://api.scnet.cn/api/llm/v1) is acceptable for your data‑handling and compliance requirements. 4) If you need stronger isolation, create an API key with limited permissions or test with non-sensitive images first. 5) The repository/homepage in metadata may be generic—if provenance is important, verify the upstream project (repository/homepage) and owner before trusting production data.
Capability Analysis
Type: OpenClaw Skill
Name: scnet-ocr
Version: 1.0.5
The scnet-ocr skill is a legitimate implementation of an OCR service wrapper for the Sugon-Scnet API (api.scnet.cn). The script main.py correctly handles file uploads, environment variable configuration for API keys, and implements standard exponential backoff for rate limiting (429 errors), with no evidence of malicious behavior, data exfiltration, or unauthorized command execution.
Capability Tags
Capability Assessment
Purpose & Capability
Name/description, SKILL.md, scripts/main.py and skill.yaml all describe an OCR client that uploads an image and returns structured JSON. Required env var SCNET_API_KEY and optional SCNET_API_BASE are appropriate and expected for a remote OCR API.
Instruction Scope
Instructions and the script operate on a user-supplied local file path and upload that file to the Scnet API. This is expected for OCR, but it means the skill will read and transmit any file path the agent is instructed to use — ensure only image/PDF files are provided. The SKILL.md does not instruct reading any unrelated local configs or secrets.
Install Mechanism
No install spec; code is a simple Python script that requires requests. No downloads from untrusted URLs or archive extraction are present in the bundle.
Credentials
Only SCNET_API_KEY (primary credential) and optional SCNET_API_BASE are requested, which align with the declared remote API usage. The script reads SCNET_API_KEY from environment or config/.env as documented and does not access other credentials or system config paths.
Persistence & Privilege
Skill does not request always: true or elevated privileges. It suggests storing an API key in a local config/.env under the skill directory (standard practice) and warns about permissions; this is within normal expectations.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install scnet-ocr - After installation, invoke the skill by name or use
/scnet-ocr - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.5
- 新增 skill.yaml 文件，完善技能元数据配置。
- SKILL.md 增加了依赖声明、环境变量说明和安全提示，明确需要 Python3 和 requests。
- 强化前置配置说明，推荐手动配置 API KEY，强调安全性，取消 AI 配置方式说明。
- 新增 required_env_vars、optional_env_vars、primary_credential、dependencies 字段说明。
- 增加依赖安装方法指引，并强调升级安全相关注意事项。
- 版本号从 1.0.2 更新至 1.0.5。
v1.0.3
scnet-ocr 1.0.3
- No file or documentation changes detected in this version.
- Functionality, usage, and API remain the same as previous release.
v1.0.2
scnet-ocr 1.0.2
- 支持的OCR识别类型由10种增加至31种，覆盖更多证件和票据类型。
- SKILL.md全面更新，详细列出所有识别能力及参数说明。
- 新增QPS速率限制说明和串行调用建议，帮助避免API限流。
- 故障排除部分新增对429（请求过多）错误的处理说明。
- 移除.clawhub/config.json示例文件，无影响核心功能。
v0.1.0
Initial release of scnet-ocr skill:
- 提供 Sugon-Scnet 通用 OCR 服务的简单封装，支持图片文字及多类证件、发票自动识别。
- 支持身份证、银行卡、营业执照、增值税发票、出租车票、火车票、航空行程单、机动车发票等 10 种类型的结构化识别。
- 输入参数明确（识别类型、图片路径），输出结构化 JSON，含置信度。
- 提供详细配置方法及常见故障排查指引。
- 支持 API Token 管理和多种图片格式。
Metadata
Frequently Asked Questions
What is Scnet Ocr?
将图片中的文字、通用文字识别, 居民身份证, 银行卡, 社保卡, 户口本, 出生医学证明, 往来港澳通行证, 往来台湾通行证, 台湾居民来往大陆通行证, 港澳居民来往内地通行证, 营业执照, 社会团体法人登记证书, 工会法人资格证书, 宗教活动场所登记证, 民办非企业单位登记证书, 事业单位法人证书, 统一社会信... It is an AI Agent Skill for Claude Code / OpenClaw, with 262 downloads so far.
How do I install Scnet Ocr?
Run "/install scnet-ocr" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Scnet Ocr free?
Yes, Scnet Ocr is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Scnet Ocr support?
Scnet Ocr is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Scnet Ocr?
It is built and maintained by SCNet-sugon (@scnet-sugon); the current version is v1.0.5.
More Skills