Back to skill

Security audit

qubrics-mailbox-agent

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill benign with medium confidence; the reviewed material suggests mailbox and calendar tasks, while one finding flags email sending beyond the stated reply purpose, and coverage is incomplete.

The verdict is benign, not a guarantee of safety, and confidence is medium. Consider whether you are comfortable with the reported ability to send standalone email; the scan was partial, with no components fully inspected and one analysis limited by a parser bound, so some evidence is incomplete.

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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (5)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
88% confidence
Finding

The description permits replies, but the supplied command sends a message using only a recipient, subject, and body. It provides no original-message or thread reference establishing reply behavior; a 'Re:' subject alone does not establish that relationship. Thus, the shown code exercises outbound message creation beyond the declared reply capability.

Content

No source excerpt is available for this finding.

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 13)May include surrounding context.

md
envVars:
      - name: "QUBRICS_ACCESS_TOKEN"
        required: false
        description: "Optional externally managed OAuth access token for CLI HTTP requests; the CLI does not refresh it."
      - name: "QUBRICS_API_KEY"
        required: false
        description: "Optional Qubrics API key or scoped agent token used by CLI, SDK, HTTP, or MCP examples."

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 28)May include surrounding context.

md
description: "Optional Qubrics CLI profile name."
      - name: "SENDMUX_ACCESS_TOKEN"
        required: false
        description: "Optional externally managed OAuth access token for CLI HTTP requests; the CLI does not refresh it."
      - name: "SENDMUX_API_KEY"
        required: false
        description: "Optional Qubrics API key or scoped agent token used by CLI, SDK, HTTP, or MCP examples."

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
80% 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 44)May include surrounding context.

md
## ClawHub account note

This ClawHub skill connects OpenClaw agents to Qubrics. Some workflows require a Qubrics account and an appropriate Qubrics API key or agent token. Qubrics account usage is external to ClawHub; do not ask users to paste secrets into chat.

Use this skill for mailbox-scoped workflows with an `smx_mbx_` key, scoped `smx_agent_` token, or REST OAuth grant with Mailbox access: read, search, triage, reply when allowed, and sync one mailbox.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
80% 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 54)May include surrounding context.

md
Before content access or mutations on any surface, validate the selected connection and verify its returned identity and authorised mailbox selection against the intended account and credential. A permission grant alone does not establish the intended identity. If identity differs or cannot be verified, stop and resolve it with the user.

- Do not ask the user to paste an API key.
- Do not use a root key for mailbox work.
- Do not create mailboxes or mailbox keys here; route those tasks to `qubrics-management`.
- Do not delete or mutate messages without explicit user confirmation.