Security audit
Security checks for vulnerabilities and agentic risk
This skill appears to be a local knowledge-base helper, but its instructions request broad command execution and persistent file-changing behavior with unclear implementation and data-flow details.
Review this before installing. It is not artifact-backed malware, but it asks for exec access, depends on a missing 'know' CLI, writes persistent files in your home directory, and recommends automated cleanup. Only use it if you understand where it will store data, what commands it will run, and whether any saved content or summaries may be sent to external services.
The activation language is overly broad and includes generic SEO-related tasks that are much wider than a simple local knowledge tool. Broad trigger wording can cause the agent to select this skill in unrelated contexts, exposing read/exec privileges and file-modifying behavior where a narrower tool should have been used.
The capability summary broadens the tool from simple knowledge capture into broad file parsing, API calls, and sandboxed command execution. In skill ecosystems, this kind of misleading summary can cause overbroad invocation and trust, making the skill more dangerous than its title and description suggest.
The capability coverage section enumerates a very broad keyword set as invocation scope without operational constraints. This increases the chance of inappropriate auto-selection and misuse of the skill's read/exec permissions in tasks outside its safe design envelope.
The skill states that add operations automatically create Markdown files and update an index, but it does not prominently warn that local files will be modified. Hidden or underemphasized side effects are risky because users may invoke the skill expecting retrieval behavior while unintentionally altering persistent local data.
The maintenance guidance recommends automated know tidy --fix execution without warning that it may bulk-reorganize files and tags. Automated reorganization can lead to unintended data changes, broken references, or loss of user expectations if triggered routinely by an agent without review.
The documentation is inconsistent about whether external APIs or LLM/API integration are required, while also claiming no extra API key is needed. This ambiguity can mislead operators about network dependencies, trust boundaries, and where data may be sent, which is a security concern for tools handling potentially sensitive knowledge content.
The skill presents itself as a narrow file-based knowledge tool, but later expands its scope to generic file processing, API integration, and command execution. This scope drift is dangerous because an agent may grant or exercise broader capabilities than a user expects, increasing the chance of unintended command execution, data access, or network use under a benign-looking label.
No suspicious patterns detected.
