1688 Item Image Optimizer

AdvisoryAudited by Static analysis on May 9, 2026.

Overview

No suspicious patterns detected.

Findings (0)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

ConcernHigh Confidence

ASI03: Identity and Privilege Abuse

What this means

If the CLI is used, the skill can persist and later use a sensitive 1688 account access key, which could affect the user's 1688 account or backend access.

Why it was flagged

The configure service can store a user-supplied 1688 access key through the local OpenClaw gateway or directly in the OpenClaw config file. This credential authority is not part of the main open-tab instructions and is not clearly declared as the skill's primary credential requirement.

Skill content

payload = {"skills": {"entries": {SKILL_NAME: {"apiKey": api_key}}}} ... skill_entry["apiKey"] = api_key

Recommendation

Do not provide an AK unless you specifically trust and need the CLI functionality. The maintainer should either remove the credential/config code or clearly declare the credential requirement, scope, storage location, and intended API use.

ConcernHigh Confidence

ASI09: Human-Agent Trust Exploitation

What this means

A user may install the skill expecting only a tab-opening helper, while the package contains broader local/API functionality and credential setup code.

Why it was flagged

The bundled CLI advertises configuration, product data, image analysis, editing, and model-generation operations, while SKILL.md presents the skill as only returning an open_tab JSON and forbids CLI use. This can mislead users about the actual capability surface included in the package.

Skill content

Commands（更多参数见项目根目录 SKILL.md）:
    configure
    get_product_title
    get_product_category
    get_product_profile
    judge_image
    edit_image
    generate_images_with_model

Recommendation

Align SKILL.md, metadata, and shipped files. Remove unused or out-of-scope CLI code, or document the CLI capabilities, credentials, and risks in the user-facing skill description.

ConcernHigh Confidence

ASI09: Human-Agent Trust Exploitation

What this means

If the CLI is run, usage metadata may be sent to the 1688 skills gateway without the user seeing that behavior in the normal skill documentation.

Why it was flagged

The tracker states that every CLI command reports a usage record to the skill gateway and silently ignores failures. This telemetry is not disclosed in the main open-tab-only skill instructions.

Skill content

每次 CLI 命令执行时，向 skill 网关上报一次调用记录，用于统计 skill 调用次数。上报失败不影响主流程，静默处理。

Recommendation

Disclose usage reporting in SKILL.md and metadata, explain what data is sent, and provide a clear opt-out or avoid telemetry for an instruction-only tab-opening skill.