Security audit
Security checks for vulnerabilities and agentic risk
This looks like a local log summarizer, but it mixes sensitive memory-file editing with unclear command, callback, API, and credential instructions.
Review before installing. Use it only with logs you are comfortable processing, require exact input and output paths, inspect the generated summary before any MEMORY.md append, and do not provide API keys, credentials, or callback URLs unless the publisher clarifies why they are needed. The package does not include the referenced Node script, so avoid running any substitute script unless you have independently reviewed it.
The display name, summary, description, and the rest of the usage instructions are written in Chinese, and the file does not offer an alternate language or state that the skill is intentionally limited to Chinese-speaking users. This may violate a language/locale policy when users are not given an explicit language choice or opt-in.
The workflow appends output into MEMORY.md, a persistent user data store, without a prominent warning or confirmation requirement. In context this is significant because the skill can permanently alter long-term memory/log artifacts, potentially polluting records, overwriting intent, or retaining distilled sensitive data longer than expected.
The example trigger phrase is overly broad and lacks clear activation boundaries, which can cause accidental invocation on unrelated user requests. In a skill with read/exec/write capabilities, accidental activation is risky because it may lead to unintended file processing or modification.
The skill claims the free version is fully local and requires no API key, but other sections instruct users to configure API keys and use external services. This inconsistency can mislead users into exposing credentials or allowing networked behavior they did not consent to, defeating the stated trust boundary of a local-only tool.
The document states async processing is unsupported while the input schema advertises a callback_url for asynchronous completion. That mismatch can cause users or agents to provide callback endpoints unnecessarily, creating confusion and possible unintended outbound requests or disclosure of internal URLs.
The core skill is described as a basic local single-file Markdown distiller, but later text expands its apparent scope to API integration and external service usage. This widens the operational expectations and can cause an agent or user to authorize capabilities beyond what was originally understood, increasing data exposure risk.
Credential handling and external API integration are documented even though they are not justified by a simple local log-processing skill. In context, this is more dangerous because the skill processes memory/log files that may contain sensitive personal or operational data, so unjustified network or credential workflows materially increase leakage risk.
No suspicious patterns detected.
