Security audit
Security checks for vulnerabilities and agentic risk
This is a straightforward email completion notifier that sends expected task details by SMTP, with no hidden persistence or destructive behavior found.
Install this only if you are comfortable sending task titles, project names, machine names, statuses, and summaries to your configured email recipients and SMTP provider. Use an app-specific SMTP password where possible, avoid secrets or sensitive client details in summaries, and skip the AGENTS.md auto-notification instruction if you want manual control per run.
The skill instructs the agent to read environment variables and potentially local files such as AGENTS.md to construct and send an email, but it does not declare any explicit tool scope or allowed-tools boundary. That omission weakens reviewability and least-privilege controls, making it easier for an agent to access more context than users may expect when preparing outbound notifications.
The skill sends task metadata externally via SMTP, including device name, project name, task title, status, and summary, but the documentation does not prominently warn that this information leaves the local environment. Users may therefore include sensitive operational details in the summary or enable the skill without understanding the external disclosure risk.
The skill directs the agent to generate and transmit a task summary after completion, creating a persistence/exfiltration channel for session-derived data. Even with the note to avoid secrets, LLM-generated summaries can inadvertently include sensitive context, and email creates a durable external record that may be retained or forwarded.
- Generate a short task title (3-8 words).
- Pick an execution status: `success`, `failed`, `partial`, `blocked`, etc.
- Write a brief result summary; avoid secrets.
Run:
No suspicious patterns detected.