xhs-prohibited-word（小红书违禁词查询）

AdvisoryAudited by Static analysis on May 10, 2026.

Overview

Detected: suspicious.insecure_tls_verification

Findings (1)

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.

NoteHigh Confidence

ASI07: Insecure Inter-Agent Communication

What this means

Text from pasted content, selected files, or extracted webpages may be transmitted to the external detection service.

Why it was flagged

The user’s submitted copy is sent to a third-party API endpoint for detection. This fits the skill’s purpose, but it means private drafts or business content leave the local environment.

Skill content

_DEFAULT_SENSITIVE_API_URL = "https://onetotenvip.com/story/cozeSkill/sensitiveWordSearch" ... payload = {"content": content, "platform": "小红书", "source": "小红书违禁词查询-ClawHub"}

Recommendation

Use the skill only with content you are comfortable sending to that service, and the maintainer should clearly disclose the endpoint and data-handling expectations.

NoteHigh Confidence

ASI07: Insecure Inter-Agent Communication

What this means

If TLS verification is turned off, submitted text could be exposed to interception or tampering on the network.

Why it was flagged

The script contains a code path that disables TLS certificate verification when ssl_verify is false, though detection defaults to verification enabled unless an environment variable disables it.

Skill content

ctx.check_hostname = False
ctx.verify_mode = ssl.CERT_NONE

Recommendation

Keep TLS verification enabled, and avoid setting XHS_SENSITIVE_WORD_VERIFY_SSL to 0/false/no unless there is a controlled and well-understood reason.

NoteMedium Confidence

ASI02: Tool Misuse and Exploitation

What this means

Accidentally providing the wrong file path or URL could process unintended content; unsafe shell quoting could also be risky in some agent runtimes.

Why it was flagged

The workflow passes user-controlled text, file paths, and URLs into a local script. This is central to the skill, but it should be done with safe argument passing and only for user-intended files or URLs.

Skill content

直接调用 `python scripts/check_sensitive_words.py --content="文案内容"` ... `--file=/path/to/file.txt --extract-only` ... `--url=https://example.com --extract-only`

Recommendation

Provide only intended files/URLs, and the runtime or maintainer should invoke the Python script with structured arguments rather than unsafe shell string interpolation.

NoteHigh Confidence

ASI04: Agentic Supply Chain Vulnerabilities

What this means

The skill may require extra package/browser installation before it works, which adds normal dependency provenance considerations.

Why it was flagged

The skill depends on third-party Python packages and a Playwright browser download, while the registry install section says there is no install spec.

Skill content

dependency:
  python:
    - python-docx==1.1.0
    - beautifulsoup4==4.12.3
    - playwright==1.58.0
  system:
    - playwright install chromium

Recommendation

Install dependencies from trusted package repositories in an isolated environment, and document the setup steps in registry metadata if possible.

Findings (1)

warn

suspicious.insecure_tls_verification

Location: scripts/check_sensitive_words.py:34
Finding: HTTPS certificate verification is disabled.