Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed push-notification helper that sends user-selected task summaries through a configured MCP server, with no hidden code or destructive behavior found.
Install this only if you want your agent to send summaries or JSON payloads to the configured push service and downstream channels. Treat the channel code like a credential, verify the destination channel/webhook, and avoid sending secrets, personal data, proprietary content, or full logs unless you have intentionally reviewed and approved that content.
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
.vscode/
# local env files
.env
.env*.local
The README explicitly promotes sending AI-collected content as JSON to external webhook/business services, but it does not provide a clear, prominent warning that task data may include sensitive information and will be disclosed to third-party endpoints. In the context of an agent skill designed to summarize and forward task results, this omission increases the risk of unintended exfiltration of secrets, internal data, or personal information.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
The usage section encourages agents to automatically push a notification after task completion, but it does not prominently warn that the resulting summaries and links may be sent to external notification providers such as browser push, chat platforms, email, or webhook receivers. Because this skill's entire purpose is outbound delivery, the lack of a strong disclosure warning makes accidental leakage of sensitive task output more likely.
No suspicious patterns detected.