ClawHub
Back to skill

Security audit

Jira

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent Jira connector, but it should go to Review because its setup instructions can run unverified installer scripts from the internet.

Install only if you trust OOMOL with access to the connected Jira account. Avoid running the pasted installer one-liners blindly; prefer a reviewed official installation path with verification where possible. Confirm any comment or issue-creation payload before allowing the skill to run write actions.

A.I.G

Vulnerability Patterns
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Findings (1)

T03 · Remote Payload Retrieval and Execution

Error
Location
SKILL.md:58
Finding

Unverified Remote Installer Download and Immediate Execution

Content
View full analysis
): ```bash curl -fsSL https://cli.oomol.com/install.sh | bash # macOS / Linux ``` ```powershell irm https://cli.oomol.com/install.ps1 | iex # Windows PowerShell ``` ``` ### Technical Analysis The installation instructions download executable content from `cli.oomol.com` and pass it directly to a command interpreter: - On macOS and Linux, `curl` sends the response body directly to `bash`. - On Windows, `Invoke-RestMethod` downloads the response body and `Invoke-Expression` executes it. The effective installer payload is controlled by the remote server and can change after the Skill has been reviewed. HTTPS protects transport integrity but does not establish that future server content is trustworthy. The instructions do not pin an installer version, verify a cryptographic signature or digest, save the script for inspection, or constrain the script's execution privileges. Installation is presented only as a fallback for systems where `oo` is missing, rather than as part of ordinary Jira operations. Moreover, the Skill declares `allowed-tools: [Bash(oo *)]`, which does not authorize `curl`, `bash`, or the PowerShell commands shown here. A runtime that strictly enforces this allowlist should block the installation commands, reducing exploitability. In an environment that does not enforce the allowlist, or if a user manually follows the instructions, the commands create an arbitrary remote-code execution channel that exceeds the privileges required merely to invoke an already-installed `oo` CLI. ### Attack Path 1. The `oo` command is unavailable, or an attacker causes the action to fail in a way interpreted as `oo: command not found`. ...[truncated 1509 chars]
Remediation
View remediation
oo-install.sh" | sha256sum --check less oo-install.sh bash oo-install.sh ``` The release URL and digest must correspond to an authentic, immutable, vendor-published release. Merely splitting download and execution without verification is insufficient. ]]>

SkillSpector

Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

External Script Fetching

High
Category
Supply Chain
Confidence
97% confidence
Finding

The skill includes a shell one-liner that downloads and immediately executes a remote script via curl ... | bash, which is a classic supply-chain and remote code execution risk. If the URL, hosting, transport, or upstream content is compromised, the agent or user environment could execute attacker-controlled code with local privileges.

Content

Scanner excerpt · SKILL.md (reported line 61)May include surrounding context.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill description says to use this skill for ANY Jira request and whenever a task involves Jira, which can cause over-broad routing and automatic invocation on incidental Jira mentions. In an agent setting, this increases the chance of unnecessary access to connected Jira data or accidental write-capable tool use when a narrower or safer path would be more appropriate.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.