ClawHub
Back to skill

Security audit

Wit.ai

Security checks for vulnerabilities and agentic risk

Overview

The skill is coherent for managing Wit.ai through OOMOL, but its setup instructions include unverified remote installer commands that can execute code locally.

Review the setup path before installing. Prefer installing the oo CLI from verified official release or package-manager instructions, and do not run the curl-to-bash or PowerShell iex commands unless you trust the OOMOL installer source and understand it can execute local code. For normal use, confirm all write or delete Wit.ai actions before allowing them.

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:74
Finding

Unverified Remote Installation Scripts Executed Directly by Shells

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 74–78
Vulnerability Type: Remote payload retrieval and execution
Risk Level: High

Vulnerable code:

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 first-time setup instructions pipe remotely retrieved content directly into Bash or PowerShell. The payload is fetched from a mutable external URL and executed without version pinning, cryptographic signature verification, checksum validation, or an opportunity to inspect the downloaded file.

Although the download domain is associated with the declared OOMOL service and installation is only suggested after a command-not-found error, these controls do not establish the integrity of the payload at execution time. Compromise of the hosting service, DNS or delivery infrastructure, publishing account, or installer itself would allow the effective code to change after the Skill has been reviewed.

Direct shell execution is not the minimum privilege necessary to explain how to install a prerequisite. A safer installation workflow can download a fixed release artifact, authenticate its provenance, and require explicit approval before execution.

Attack Path

  1. The oo command is unavailable, causing the agent or user to consult the first-time setup instructions.
  2. An attacker compromises the installer endpoint, its publishing process, or another part of the delivery chain.
  3. The attacker replaces the remotely served installer with a malicious shell or PowerShell payload.
  4. The documented curl | bash or irm | iex command retrieves the changed payload.
  5. Bash or PowerShell executes it immediately without integrity verification or review.
  6. The payload performs arbitrary operations using the privileges and accessible resources of the invoking ...[truncated 758 chars]
Remediation
View remediation

Remediation Suggestions

  • Remove the direct curl | bash and irm | iex execution patterns.
  • Prefer an operating-system package manager or an official package repository with signed metadata and pinned package versions.
  • If a standalone installer is required, reference a version-specific release artifact rather than a mutable generic installation URL.
  • Download the artifact to a local file without executing it immediately.
  • Publish and verify a cryptographic signature using a trusted, independently distributed signing key. At minimum, verify a release-specific SHA-256 checksum obtained through a separate authenticated channel.
  • Allow the downloaded script to be reviewed before execution.
  • Require explicit user approval before installing software or executing an installer.
  • Run installation with ordinary user privileges unless elevated privileges are demonstrably required, and clearly disclose any requested elevation.
  • Document the expected files, directories, network destinations, and configuration changes made by the installer.
  • Fail closed if signature or checksum verification fails; do not automatically fall back to unverified execution.

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
98% confidence
Finding

The skill instructs users to install the CLI via a remote script piped directly into a shell, which executes network-fetched code without verification. If the install endpoint, transport path, or hosting account is compromised, this becomes an immediate arbitrary code execution path on the user's machine.

Content

Scanner excerpt · SKILL.md (reported line 74)May include surrounding context.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The manifest description uses expansive activation language like "ANY Wit.ai request" and "Whenever a task involves Wit.ai," which is broader than a narrowly scoped trigger phrase. This can cause unintended invocation for loosely related Wit.ai tasks because it does not define boundaries, exclusions, or concrete trigger examples.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.