Back to skill
Skillv1.0.0

ClawScan security

DingJi Long Image · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignApr 8, 2026, 1:48 PM
Verdict
benign
Confidence
high
Model
gpt-5-mini
Summary
An instruction-only skill that describes how to assemble and render long-form course images; its declared inputs, runtime steps, and outputs are coherent with the stated purpose and it does not request credentials or unusual system access.
Guidance
This is an instruction-only skill that outlines how to build and export long-form math lesson images; it asks for no credentials and has no install step. Two practical checks before enabling: (1) confirm your platform provides the referenced helper (scripts.generate_long_image) and the lightclaw_upload_file API or the calls will fail; (2) be aware generated images are saved to /root/.openclaw/media/outbound and uploaded—if you embed sensitive content in the chapter text or brand fields, it will be persisted and uploaded. If you rely on other linked skills (course-long-image), review those implementations for any code or network behavior you must trust.

Review Dimensions

Purpose & Capability
okName/description (generate long images for math chapters) matches the SKILL.md content. No unrelated binaries, environment variables, or external credentials are requested. References to branding, chapter content, LaTeX, and output/upload behavior are appropriate for an image-generation skill.
Instruction Scope
noteThe instructions include Python-snippet examples that call a helper script (scripts.generate_long_image) and instruct uploading via lightclaw_upload_file to /root/.openclaw/media/outbound. This is consistent with the described workflow but assumes platform-provided helper scripts/APIs or other skills (e.g., course-long-image). There are no instructions to read arbitrary user files, exfiltrate secrets, or contact unknown external endpoints.
Install Mechanism
okNo install spec or downloaded code is present (instruction-only). Nothing is written to disk by the skill itself at install time, so install risk is minimal.
Credentials
okThe skill declares no required environment variables or credentials. Its runtime instructions do not request tokens/keys. The only filesystem path referenced is a platform media outbound path for storing generated images, which is proportionate to the stated output behavior.
Persistence & Privilege
okalways is false and model invocation is not disabled (normal). The skill does not request persistent presence, nor does it instruct modifying other skills' configuration or accessing other skills' credentials.