Back to skill

Security audit

SMTP

Security checks for vulnerabilities and agentic risk

Overview

This SMTP skill is coherent and safety-scoped: it supports email testing and sending with explicit confirmation, local notes, and secret-handling warnings.

Install only if you are comfortable letting the agent help test or send SMTP mail. Use approved canary inboxes first, confirm every live send, provide credentials only through a trusted runtime secret source, and keep ~/smtp notes free of passwords, tokens, private keys, full recipient lists, and full message bodies unless truly necessary.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

Anti-Refusal Statement

High
Category
Anti-Refusal
Confidence
80% confidence
Finding

Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.

Content

Scanner excerpt · send-flow.md (reported line 3)May include surrounding context.

md
# Send Flow - SMTP

Use this workflow for every SMTP task so the agent does not skip safety gates.

## Step 1: Classify the job

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 142)May include surrounding context.

md
- guess live-send approval from vague user intent
- keep credentials in `~/smtp/`
- treat queue acceptance as the final delivery result
- broaden a one-off test into a bulk-send workflow without consent

## Related Skills
Install with `clawhub install <slug>` if user confirms:

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
80% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · setup.md (reported line 56)May include surrounding context.

bash
mkdir -p ~/smtp
touch ~/smtp/{memory.md,provider-profiles.md,send-log.md,deliverability-notes.md}
chmod 700 ~/smtp
chmod 600 ~/smtp/{memory.md,provider-profiles.md,send-log.md,deliverability-notes.md}

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
80% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · setup.md (reported line 57)May include surrounding context.

mkdir -p ~/smtp touch ~/smtp/{memory.md,provider-profiles.md,send-log.md,deliverability-notes.md} chmod 700 ~/smtp chmod 600 ~/smtp/{memory.md,provider-profiles.md,send-log.md,deliverability-notes.md}

text

If the files are empty:

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

The template directs creation of a persistent SMTP memory file containing operational details like provider/host names, sender-domain control, safe canary inboxes, and known-good auth/TLS combinations, but it does not warn against storing secrets such as SMTP passwords, API keys, or full recipient data. In an email infrastructure context, users may reasonably over-collect sensitive configuration material in this file, increasing the risk of credential exposure or leakage of internal mail-routing information.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.