ClawHub

multi-image-to-text

v1.0.0

批量识别图片中的文字内容并按图片分段输出为结构化文档；当用户需要从多张图片中提取文字、整理图片文字内容、将图片文字转为可编辑文档时使用

⭐ 0· 73·0 current·0 all-time
by@asiangiantduck
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name/description (batch image→text extraction) align with the SKILL.md: it instructs the agent to accept images, call a read_image OCR tool per image, and produce structured Markdown. No unrelated binaries, env vars, or config paths are requested.
ℹ
Instruction Scope
Instructions are narrowly scoped to: receive images, call read_image for each image, format and output a Markdown document, and optionally write a .md file under ./ with a timestamped name. The doc claims images are used only in the current session and not stored — this is a behavioral claim but cannot be enforced by the instruction file alone; actual persistence/processing depends on the platform/tool implementing read_image and the agent runtime.
✓
Install Mechanism
Instruction-only skill with no install spec and no code files. No downloads or package installs are performed by the skill itself.
✓
Credentials
The skill declares no environment variables or credentials. Its functionality (OCR on user-supplied images) does not require additional secrets, so requested access is proportionate.
✓
Persistence & Privilege
always is false, the skill is user-invocable, and it does not request modification of other skills or global agent settings. It may write a .md file to the local working directory when asked, which is consistent with its stated output options.
Assessment
This skill appears coherent and low-risk: it does OCR on images via the agent's read_image tool and outputs Markdown. Before installing or using it, confirm how your platform implements the read_image tool (where images are sent/processed and whether they are retained), avoid uploading highly sensitive images unless you trust the platform's data handling, and be aware that the skill can create a .md file in the agent's working directory when asked. If you need stronger guarantees about deletion or local-only processing, verify those properties with the platform or prefer local/offline OCR tools.

Like a lobster shell, security has layers — review code before you run it.

latestvk97abyb0hvz3c497jvnew4nw4d843m70

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Comments