Email Intelligence

PassAudited by VirusTotal on May 12, 2026.

Overview

Type: OpenClaw Skill Name: email-intelligence Version: 1.0.0 The skill bundle is classified as benign. The `SKILL.md` provides clear instructions and descriptions without any prompt injection attempts. The `scripts/email_classify.py` script securely executes the `himalaya` CLI tool using `subprocess.run` with a list of arguments, preventing shell injection. The script's actions are directly aligned with its stated purpose of analyzing email inbox health and do not involve data exfiltration, unauthorized remote control, persistence mechanisms, or access to unrelated sensitive files. It relies on the user having `himalaya` configured, but the script itself introduces no new vulnerabilities or malicious behavior.

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

Running the skill lets it inspect email metadata from the configured account, which may include personal or work messages.

Why it was flagged

The skill depends on an already-configured email account, which means it can act through whatever IMAP access himalaya has.

Skill content
Requires himalaya CLI configured with IMAP.
Recommendation

Use it only with the intended mailbox, review the himalaya configuration, and prefer the least-privileged/read-only setup available.

What this means

Private sender and subject information may appear in the generated report, and subject lines should not be treated as trusted instructions.

Why it was flagged

The script includes email sender addresses and subject lines in its ghost report output, placing private and potentially untrusted email metadata into the agent/report context.

Skill content
'senderEmail': sender.get('addr', ''),
                'subject': email.get('subject', ''),
Recommendation

Keep reports local, redact sensitive senders or subjects before sharing, and treat email-derived text as untrusted content.

What this means

Users may not realize before reading the skill instructions that it depends on an external mail CLI and an IMAP-configured account.

Why it was flagged

The skill documents runtime dependencies that are not reflected in the registry metadata, which lists no required binaries, environment variables, or primary credential.

Skill content
- **himalaya CLI** configured with IMAP access
- Python 3.6+
Recommendation

Declare himalaya, Python, and the email-account dependency in metadata or installation requirements for clearer user review.