抖音自动回复助手

SuspiciousAudited by ClawScan on May 10, 2026.

Overview

The skill is coherent and not obviously malicious, but it asks for a full Douyin session cookie and is designed to run unattended actions on the user's Douyin account.

Install only if you are comfortable giving the tool a Douyin session cookie and letting it act on your account. Prefer a test or dedicated account, protect config.json and logs, set low limits, review keywords carefully, and understand Douyin platform-rule and account-ban risks.

Findings (5)

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 full session cookie can let the bot act as the user's Douyin account; if the file is exposed or the bot misbehaves, the account could post unwanted replies, send unwanted DMs, or face platform restrictions.

Why it was flagged

The skill asks users to copy a full Douyin browser session cookie into local configuration. This is high-impact account authority and is under-declared by the registry metadata, which lists no primary credential.

Skill content
复制 cookie 字段的全部内容 ... python scripts/config_manager.py cookie "粘贴你刚才复制的 cookie"
Recommendation

Use official scoped OAuth/API credentials if available, declare the credential requirement, store secrets securely, warn users about plaintext config.json, and recommend a dedicated/test account rather than a main account.

What this means

Incorrect keywords or unexpected comments could trigger public replies or DMs at scale, potentially annoying users, damaging reputation, or triggering Douyin anti-spam/account controls.

Why it was flagged

The bot is designed to continuously process comments and send replies/private messages after the user starts it. It has delay and daily-limit controls, but no per-action confirmation or clearly bounded target scope.

Skill content
while True: ... comments = self.get_comments() ... self.process_comment(comment) ... self.send_private_message(user_id, pm_message)
Recommendation

Add dry-run mode, per-message review options, clear video/account scoping, conservative defaults, and a visible stop/rollback procedure.

What this means

Users may treat the setup as safer than it is and paste a main-account cookie without protecting local files.

Why it was flagged

The privacy wording may cause users to underestimate the sensitivity of copying a full session cookie and creating local logs/config files, even though no external exfiltration is shown.

Skill content
用户隐私:不收集用户个人信息;不存储聊天记录;本地运行,数据安全
Recommendation

Clarify exactly what is stored locally, where credentials/logs live, and what risks remain even without external data collection.

What this means

Someone with access to the skill directory could read account activity details or infer private-message targets/content.

Why it was flagged

The bot writes persistent local logs that can include user IDs and outgoing message content. No external sharing is shown, but the files may contain sensitive account-interaction data.

Skill content
logging.FileHandler('douyin_bot.log') ... logger.info(f"发送私信给用户 {user_id}: {message}")
Recommendation

Document log contents, avoid logging sensitive identifiers/messages by default, and provide log rotation/redaction/cleanup guidance.

What this means

Future installs may receive different dependency versions, which can affect reliability or security.

Why it was flagged

The dependency is unpinned. This is expected for a simple Python HTTP integration, but the exact installed version can vary over time.

Skill content
requests>=2.28.0
Recommendation

Pin or constrain dependencies more tightly and provide a lockfile or verified install instructions.