ClawHub
Back to skill

Security audit

agentpmt-workflow-creator

Security checks across malware telemetry and agentic risk

Overview

This skill is coherent for AgentPMT workflow management, but it gives agents remote authority to publish, delete, and restore workflows without strong confirmation guidance.

Install only if you want an agent to manage AgentPMT workflow drafts on your behalf. Before allowing actions like publish, delete, restore, visibility changes, or context attachment, require the agent to show the target workflow ID, current draft revision, and intended effect, then get explicit confirmation.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Vague Triggers

Medium
Confidence
89% confidence
Finding
The activation keywords are broad and generic enough to match many ordinary automation requests, which can cause the skill to be invoked outside the user's explicit intent. Because this skill performs remote workflow-management actions against an external service, unintended invocation could lead to undesired draft creation, updates, publishing, or other side effects.

Missing User Warnings

Medium
Confidence
84% confidence
Finding
The skill exposes a destructive `delete` capability but does not prominently warn about data loss, ownership checks, or irreversibility before presenting it as a normal core action. In a workflow-authoring context, accidental or ambiguous use could delete drafts or work product a user expected to preserve.

Missing User Warnings

Medium
Confidence
87% confidence
Finding
The schema documents destructive state-changing actions such as delete, publish, update, attach/detach, and restore, but it does not include clear user-facing warnings or confirmation guidance about their effects. In an agent-executed workflow builder, this increases the chance of unintended mutations or deletion if an upstream prompt, misinterpretation, or malicious instruction causes the agent to invoke these actions automatically.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The restore_published and restore_version actions state that they restore snapshots into the current draft, but they omit a prominent warning that current draft contents may be overwritten. In this skill context, agents are explicitly empowered to manage workflow drafts remotely, so missing overwrite warnings materially raises the risk of accidental loss of in-progress work or rollback to attacker-selected versions.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.