email skill

PassAudited by VirusTotal on May 12, 2026.

Overview

Type: OpenClaw Skill Name: mailbox Version: 1.0.0 The skill bundle describes an OpenClaw skill to manage email using a `mailbox-cli`. All instructions and examples in `SKILL.md` are directly related to the stated purpose, including installing the CLI via `npm` and accessing `~/.config/mailbox/auth.json` for credentials. There is no evidence of prompt injection attempts against the agent, data exfiltration, malicious execution beyond standard package installation, persistence mechanisms, or obfuscation. Safety rules for destructive operations are explicitly mentioned, indicating a focus on secure usage.

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

If installed and used, the agent may be able to read mailbox contents and perform mailbox actions through the configured account.

Why it was flagged

The skill relies on a local mailbox credential profile, while the registry declares no primary credential or required config path. Email account credentials are high-impact because they can expose private messages and enable account actions.

Skill content
- Credentials in `~/.config/mailbox/auth.json`
Recommendation

Use a dedicated or least-privilege mailbox account if possible, verify what is stored in the auth file, and require explicit user approval before account-changing actions.

What this means

A mistaken or over-broad delete command could remove email from the mailbox.

Why it was flagged

The skill documents a destructive delete operation. It is aligned with the stated email-management purpose and includes safety rules, but deletion is still a high-impact action.

Skill content
`mailbox email delete <email_uid> --account-id <account_id> --folder INBOX --confirm --json`
Recommendation

Only allow deletion after reviewing the target account, folder, and email UID; prefer dry-run behavior before using --confirm.

What this means

Installing or using the wrong or compromised CLI package could expose email credentials or perform unintended mailbox actions.

Why it was flagged

The skill depends on a globally installed external npm CLI that is not included in the reviewed artifact set. Because that CLI will handle mailbox credentials and email actions, package provenance and version pinning matter.

Skill content
- mailbox CLI installed (`npm install -g mailbox-cli`)
Recommendation

Verify the npm package source, maintainer, version, and integrity before installing; prefer pinned versions and a trusted project homepage or repository.

NoteMedium Confidence
ASI10: Rogue Agents
What this means

Mailbox access could continue through scheduled or repeated monitoring workflows if enabled.

Why it was flagged

The skill includes a monitoring command and says OpenClaw handles scheduling. This is consistent with email monitoring, but it may create recurring or long-running access to mailbox data depending on how it is invoked.

Skill content
`mailbox monitor run --json`
Recommendation

Confirm how monitoring is started, stopped, and scheduled before enabling it.