Back to skill

Security audit

qubrics-getting-started

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious because it reports a mismatch between the onboarding purpose and management access, plus an install command that is not pinned to a reviewed version.

Consider the reported mismatch and unpinned installation source before installing. The scan is partial: it inspected one Markdown component but could not resolve several references, so the evidence is incomplete. The verdict is suspicious (Review), not a finding that the author is malicious.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (7)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The code primarily configures root-authenticated management access and performs management queries. It contains no inbox registration, owner linking, or runtime selection. Although profile configuration and a read-only request could support onboarding, reliance on an existing root credential and access to management resources materially differ from the stated keyless agent-onboarding purpose. No data exfiltration or destructive operation is shown.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
90% confidence
Finding

The supplied chunk is an authenticated mailbox-listing operation, not an onboarding implementation. Although listing a mailbox could qualify as a harmless initial call, the chunk requires an existing key and contains none of the registration, owner-linking, or runtime-selection behavior central to the declared purpose.

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 31)May include surrounding context.

md
description: "Example secret-store variable for a root key; explicitly pass its value through QUBRICS_API_KEY or the authorised request header."
      - 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 47)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 to get a user from "I have a Qubrics task" to the correct surface, key kind, package, and first verified call.

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 57)May include surrounding context.

md
gistration flow, complete registration first, then verify the newly saved profile before mailbox access, sending or other task work. For existing credentials or profiles, before task work on any surface, validate the selected connection and verify its returned identity and any 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 print API keys.
- Prefer the intended permission-restricted local CLI profile, or load only the intended credential from the user's secret manager into the isolated child environment after clearing inherited overrides. Authorised profile and secret-store files are expected credential storage, not a leak.
- For self-registration, let the CLI create and persist the agent profile; never copy its raw credential into chat, logs, prompts, screenshots,

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
93% confidence
Finding

The command npx skills add qubrics/skills executes an unpinned package and installs skill content from an unpinned repository reference, allowing upstream changes to alter what the agent runs or trusts. This is a supply-chain reproducibility risk rather than an unpinned MCP server specifically; isolated execution and credential clearing reduce exposure but do not establish dependency integrity.

Content

No source excerpt is available for this finding.