Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed micro-learning helper that uses conversation context, web search, and a saved HTML output in ways that match its stated purpose.
Install if you want a proactive learning-card feed while waiting on coding-agent work. Be aware that it may use current conversation context to form web searches and will save a browser-viewable HTML file, so avoid using it in chats containing secrets, confidential project details, or sensitive internal topics unless you are comfortable with that content being reflected in the generated page.
The trigger conditions are extremely broad and include generic phrases like 'I'm waiting' plus inferred idle states, which can cause the skill to activate without clear user intent. Because the skill performs web searches and generates artifacts/files, unintended invocation can lead to unnecessary external access, confusing behavior, and surprise side effects.
The README says the skill generates a standalone HTML page and provides a default filesystem output path, but it does not warn users that conversation-derived content will be written to disk. Because the skill is context-aware and uses current task context, the generated file may persist sensitive project details, internal URLs, or proprietary topics in a shareable artifact. The danger is amplified by the stated browser-viewing, bookmarking, and sharing use case, which encourages persistence and redistribution of potentially sensitive data.
The trigger list includes broad everyday phrases such as 'learn something', 'knowledge cards', and 'feed me something', and the README also encourages proactive activation when the user is merely inferred to be idle. That can cause the skill to run without sufficiently clear user intent, leading to unsolicited context processing and output generation in unrelated conversations. In a skill that inspects current task context and produces artifacts, over-broad activation increases the risk of privacy leakage and unwanted side effects.
The skill mandates writing a standalone HTML file into user-data and presenting it, but does not clearly disclose the write before doing it or obtain consent. Silent file creation is a risky side effect because it can surprise users, create unwanted persistent artifacts, and normalize hidden filesystem writes by skills.
No suspicious patterns detected.
