Douyin Video Search

PassAudited by ClawScan on May 10, 2026.

Overview

The skill appears to perform the advertised Douyin search through TikHub, but users should notice that it needs a TikHub API token and sends search requests to TikHub.

Before installing, make sure you are comfortable giving this skill a TikHub API token and sending your Douyin search terms to TikHub. The code shown is consistent with the advertised search function, but the publisher should clean up debug output and improve source/dependency metadata.

Findings (4)

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.

Low

#ASI03: Identity and Privilege Abuse

What this means

The agent can make TikHub API requests using the user's token when this skill is invoked.

Why it was flagged

The skill requires a provider API token stored in the OpenClaw config. This is purpose-aligned for TikHub API search, but it grants the skill delegated access to the user's TikHub account/API quota.

Skill content

本技能需要一个有效的 TikHub API Token 才能工作。请在您的 `~/.openclaw/config.json` 文件中添加以下配置项。 ... "tikhub_api_token": "YOUR_TIKHUB_API_TOKEN"

Recommendation

Use a revocable TikHub token with the minimum needed scope or quota, and rotate it if you no longer trust the skill or environment.

Low

#ASI07: Insecure Inter-Agent Communication

What this means

TikHub receives the search keywords, filters, pagination values, and the API token used for the request.

Why it was flagged

The script sends the search payload and bearer-token authorization header to the TikHub API endpoint. This external provider call is disclosed and central to the skill's purpose.

Skill content

TIKHUB_SEARCH_URL = "https://api.tikhub.io/api/v1/douyin/search/fetch_general_search_v1" ... resp = requests.post(TIKHUB_SEARCH_URL, json=payload, headers=headers, timeout=30)

Recommendation

Avoid searching highly sensitive terms unless you are comfortable sharing them with TikHub, and review TikHub's privacy and token-management practices.

Info

#ASI09: Human-Agent Trust Exploitation

What this means

The agent transcript may include extra raw response fragments or debug lines, which can clutter output or expose more provider response detail than expected.

Why it was flagged

The script prints debug status and the first 500 characters of the raw provider response on every request, even though the documentation presents raw JSON output as controlled by the --raw flag.

Skill content

print(f"DEBUG: status_code={resp.status_code}")
print(f"DEBUG: response_text={resp.text[:500]}")

Recommendation

Remove unconditional debug prints or gate them behind an explicit --debug option.

Info

#ASI04: Agentic Supply Chain Vulnerabilities

What this means

It may be harder to verify the skill's origin or reproduce the exact dependency environment if setup is done manually.

Why it was flagged

The source URL appears to be a placeholder and the dependency is unpinned. The runnable script is included and no automatic installer is shown, so this is a provenance/reproducibility note rather than a malicious indicator.

Skill content

source: https://github.com/your-repo/douyin-video-search ... requirements:
    - python
    - "pip:requests"

Recommendation

Publish accurate source metadata and pin or document dependency versions for more reproducible installation.