Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Aws Wechat Article Images

v1.0.0

为公众号文章生成封面图和正文配图,根据文章内容自动匹配风格。当用户提到「封面」「配图」「插图」「生成图片」「给文章加图」「做个封面」「文章插图」「配个图」时使用。

0· 31·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Pending
View report →
OpenClawOpenClaw
Suspicious
medium confidence
Purpose & Capability
The skill's name and description (generate cover/body images for WeChat articles) match the included script and SKILL.md workflow. The script calls external image-generation APIs using settings in .aws-article/config.yaml and an API key from aws.env, which is coherent with the stated purpose. However, the registry metadata lists no required environment variables even though the runtime requires IMAGE_MODEL_API_KEY in aws.env; that documentation/metadata omission is an inconsistency.
!
Instruction Scope
SKILL.md explicitly instructs the agent to read and write repository files: .aws-article/config.yaml, aws.env (repo root), article.md/article.yaml, imgs/ and possibly article.html. It also instructs running the bundled Python script to call external model endpoints. Reading aws.env (a dotenv file at repo root) is broader than a single declared key and could expose unrelated secrets. Otherwise the file reads/writes and network calls are directly related to generating and inserting images.
Install Mechanism
No install spec; this is an instruction-only skill with one Python script that uses standard library modules (urllib, yaml). Nothing is downloaded or executed from untrusted URLs in the provided files; no package install step is required.
!
Credentials
At runtime the script expects an IMAGE_MODEL_API_KEY stored in a repository aws.env and an image_model section in .aws-article/config.yaml. The registry metadata did not declare required env vars. The script parses the whole aws.env file (yielding all keys), so if aws.env contains other secrets those would be read by the skill. Requesting/reading repository-wide dotenv without declaring or limiting which keys are required is disproportionate and worth confirming before use.
Persistence & Privilege
Skill flags in registry: always=false and user-invocable=true (normal). skill.json grants filesystem, network, and shell permissions — which align with the SKILL.md (running python, reading/writing article files, calling network APIs). The shell permission is broad but appears intended to allow the agent to run the python script; still, granting shell access increases risk if the agent is allowed to run arbitrary commands beyond the described ones.
What to consider before installing
This skill appears to implement the promised image-generation workflow, but review and confirm before installing: - Inspect aws.env in the repository root. The script expects IMAGE_MODEL_API_KEY there; ensure aws.env does not contain other unrelated secrets (database credentials, cloud keys). If it does, move the image model key to a dedicated minimal-scope secret or declare the required var explicitly. - Verify .aws-article/config.yaml contents and the model provider/base_url to ensure the endpoint is trusted. The script will call whatever base_url and API key are configured. - Review scripts/image_create.py (it is included) to be sure you accept its network calls and any file writes (imgs/, article.html modifications when triggered). The script reads the whole dotenv file but only uses IMAGE_MODEL_API_KEY — nevertheless, reading the file is broad. - Note skill.json requests filesystem, network, and shell permissions. Only allow this skill in agents you trust and where running the specified python commands is acceptable. - To reduce risk: declare IMAGE_MODEL_API_KEY as a required env var in the registry or store only that key in aws.env; restrict the API key to least privilege and an appropriate billing/account; test the script using a non-production key. What would change this assessment to benign: registry metadata that documents the required IMAGE_MODEL_API_KEY and shows a trusted homepage/owner, and assurance that aws.env contains only the image API key (or the code was changed to read only the single required var rather than the entire dotenv). Conversely, if additional hidden network endpoints or code that exfiltrates other secrets were found, risk would be higher.

Like a lobster shell, security has layers — review code before you run it.

latestvk97ebfsb6m6q9tf04chn3mevp1841ec4

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments