ClawHub
Back to skill

Security audit

Jira Api Toolkit Free

Security checks across malware telemetry and agentic risk

Overview

This Jira skill is mostly a read-only integration, but its metadata and instructions grant or describe broader write-style authority than the free read-only purpose supports.

Install only if you are comfortable granting a CLI-backed agent access to your Jira Cloud account. Treat it as a Review item because the free edition claims read-only behavior but requests or describes broader write-style capability; use least-privilege Jira access, avoid broad JQL queries, and do not expose API keys, cloud IDs, whoami output, or project inventory unnecessarily.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
94% confidence
Finding
The skill repeatedly claims to be read-only, but the capability sections explicitly say the agent can use parameters that support create, modify, import, export, save, and reset operations. This inconsistency can mislead an agent or user into permitting broader actions than intended, increasing the risk of unintended state changes, data exfiltration, or unsafe execution paths when the skill is invoked.

Intent-Code Divergence

Medium
Confidence
93% confidence
Finding
The inline capability text contradicts the earlier read-only promise by describing support for create/modify/import/save-style actions. In agent ecosystems, ambiguous skill contracts are dangerous because planners may rely on the broader wording and route sensitive or state-changing tasks to a skill that was expected to be non-destructive.

Vague Triggers

Medium
Confidence
83% confidence
Finding
The trigger conditions are overly broad, covering generic API integration, interface docking, webhook configuration, and system connection scenarios. Such vague activation criteria can cause the skill to be selected for unrelated tasks, exposing Jira OAuth context, prompting unnecessary exec usage, or causing accidental disclosure of project metadata in situations where the user did not intend to access Jira.

Missing User Warnings

Medium
Confidence
86% confidence
Finding
The documentation instructs users to log in, create OAuth connections, query identity, and enumerate accessible Jira resources, but it does not clearly warn that these actions involve transmitting authentication material and potentially sensitive account or tenant metadata. In an agent setting, missing disclosure and handling guidance can lead to oversharing of cloud IDs, account identity, project inventory, or auth context into logs, prompts, or downstream tools.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.