Gmail Tool

PassAudited by VirusTotal on May 11, 2026.

Overview

Type: OpenClaw Skill Name: gmail-tool Version: 1.0.0 The skill bundle contains metadata and documentation for a Gmail CLI utility. The instructions in SKILL.md describe standard, legitimate usage of SMTP and IMAP protocols via environment variables and Google App Passwords. No malicious code, suspicious URLs, or prompt injection attempts were identified in the provided files (_meta.json and SKILL.md).

Findings (0)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

Anyone or any process that can access these environment variables may be able to read from or send mail through the configured Gmail account.

Why it was flagged

The skill asks the user to provide a Gmail address and Gmail App Password, which grants SMTP/IMAP access to the account. This is expected for the stated Gmail send/read purpose, but it is sensitive account authority.

Skill content
export GMAIL_USER="your-email@gmail.com"
export GMAIL_APP_PASSWORD="xxxxxxxxxxxxxxxx"
Recommendation

Use a dedicated Gmail account if possible, protect the environment variables, revoke the app password when no longer needed, and avoid using this with a primary or highly sensitive mailbox.

What this means

A mistaken or overly autonomous invocation could expose inbox contents to the agent context or send an unintended email.

Why it was flagged

The documented CLI can send outbound email and read recent inbox messages. These actions are aligned with the skill purpose, but they are impactful and should remain user-directed.

Skill content
./gmail-tool send "recipient@example.com" "Subject" "Body text"
./gmail-tool read 10
Recommendation

Require explicit confirmation before sending emails, keep read counts small, and verify recipient, subject, and body before execution.

What this means

The installed skill may not work as documented, and any separately obtained executable would need its own trust and security review.

Why it was flagged

The package contains only documentation and metadata, while SKILL.md instructs users to run a gmail-tool executable. The actual CLI implementation is not present for review.

Skill content
2 file(s): SKILL.md (2279 bytes); _meta.json (25 bytes)
Recommendation

Only run an implementation from a trusted, reviewed source; do not supply a Gmail App Password to an unverified script or binary.

What this means

If configured, the task could keep running on a schedule and send emails based on inbox content without further prompts.

Why it was flagged

The documentation includes an optional cron-job example that would repeatedly monitor email and send alerts. This is disclosed and purpose-aligned, but it is persistent automation if the user installs it.

Skill content
0 9 * * 1-5 ./gmail-tool read 1 | grep -q "Important" && ./gmail-tool send "you@example.com" "Check Email" "Found important email"
Recommendation

Only add cron jobs intentionally, review the trigger logic carefully, and remove scheduled tasks when no longer needed.