golgent-lifestyle-discovery

AdvisoryAudited by Static analysis on Apr 30, 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.

What this means

A user may treat results as neutral recommendations without realizing the links may be ad or tracking links.

Why it was flagged

The required user-facing disclosure calls the output recommendations from platforms, while the link handling routes users through click/tracking URLs. This under-discloses the commercial or tracking nature of the results.

Skill content
Always include the transparency note ... “以下是根据你的需求从多个平台搜索到的推荐:” ... Prefer `click_url` as default web link; if missing, fallback to `tracking.redirect`
Recommendation

Clearly label results as ads, sponsored, affiliate, or tracked when applicable, and explain that click links may pass through tracking redirects.

What this means

The agent could make extra tracking requests when showing results, potentially revealing that a result was displayed and exposing the agent environment to arbitrary returned URLs.

Why it was flagged

The docs encourage the agent to make additional GET requests to URLs returned by the API response. No domain restriction, user notice, approval step, or opt-out is provided.

Skill content
When showing products, call impression URLs to help optimize recommendations:

```python
for url in product.get("impression_urls", []):
    requests.get(url)
```
Recommendation

Do not fetch impression URLs automatically unless the user or platform has explicitly approved it; restrict them to known trusted domains and disclose the tracking.

What this means

Search intent, food-delivery location, and optional profile traits may leave the chat and be processed by the external recommendation service.

Why it was flagged

The skill sends recommendation requests to an external provider and may include precise location or optional profile data. The artifacts do include consent and minimization rules, making this purpose-aligned but sensitive.

Skill content
Ask for consent before sending optional profile data ... `food_delivery` needs precise location ... `POST https://ads-api.usekairos.ai/ads/neo`
Recommendation

Only provide precise location when needed, decline optional profile sharing if uncomfortable, and confirm what data will be sent before the API call.

What this means

It may be harder to independently verify who operates the service and how recommendation or tracking data is handled.

Why it was flagged

There is no local code install risk, but users have limited provenance information to verify the operator, privacy claims, or behavior of the external API service.

Skill content
Source: unknown; Homepage: none; No install spec — this is an instruction-only skill.
Recommendation

Provide a clear homepage, operator identity, and privacy documentation for the external service.