Security audit
Security checks for vulnerabilities and agentic risk
This skill is a coherent local Hugo blog publishing helper, with disclosed file-writing behavior users should review before using on existing posts.
Before installing, confirm this is intended for your Hugo blog workflow. Use it on a version-controlled blog directory, check the generated destination path before publishing, and be aware that running it can overwrite an existing post and replace the source document's front matter in the generated output.
The skill documents capabilities to read environment variables and read/write arbitrary local files, but it does not declare any explicit tool scope or permission boundaries. In an agent setting, this increases the chance the skill is invoked with broader file or env access than users expect, enabling unintended access to local content or writes outside the intended Hugo workspace.
The trigger phrases are broad enough to match common publishing or blog-related requests, which can cause the skill to activate unexpectedly in contexts where the user did not intend file-moving or publication actions. Because the skill performs write operations and may overwrite content, overbroad activation materially raises the risk of unintended state-changing actions.
The documentation states that existing target files will be overwritten, but the workflow does not prominently require an explicit confirmation immediately before publication. In a skill that writes into a content repository, silent or under-warned overwrite behavior can destroy existing posts or replace trusted content with unintended data.
The skill notes that original front matter will be removed, but this destructive transformation is not highlighted as a major safety checkpoint in the publish flow. Removing front matter can strip important metadata such as draft status, aliases, taxonomy, canonical URLs, or custom rendering settings, causing publication mistakes or data loss.
This shell example file presents all human-readable guidance in Chinese, including setup and invocation instructions. That creates a language-specific usability policy issue because the skill does not offer an alternative language or indicate that the locale restriction is intentional and justified.
This is a natural-language policy issue because the file presents its interface and documentation in one language only, which can exclude users who do not understand Chinese. The file does not offer an opt-in language choice or explain that the tool is intentionally limited to a Chinese-only audience or region.
The script echoes raw environment variable values, including local filesystem paths, to stdout when --check-config is used. While this is not a severe issue in a local CLI context, it can leak sensitive workstation or repository layout information into terminal logs, CI logs, screenshots, or shared session transcripts.
No suspicious patterns detected.
