ClawHub
Back to skill

Security audit

Ocli Api

Security checks across malware telemetry and agentic risk

Overview

This is a legitimate API helper, but it can give an agent broad authenticated control over arbitrary APIs without enough built-in scoping or safety guidance.

Install only if you are comfortable giving the agent shell-mediated access to the APIs you configure. Use trusted OpenAPI specs and base URLs, create least-privilege tokens, prefer test or read-only profiles for exploration, and require explicit human approval before commands that create, update, delete, publish, or otherwise change remote data.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
87% confidence
Finding
The manifest description is extremely broad and positions the skill as able to turn any OpenAPI/Swagger API into callable CLI commands without stating usage boundaries or approval constraints. In an agent setting, that broad activation scope increases the chance the skill is invoked for sensitive, destructive, or unreviewed API actions, especially when paired with shell execution and dynamic endpoint discovery.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill explicitly instructs use of bearer tokens and execution of arbitrary API calls to external services, but it provides no warning about data exfiltration, credential sensitivity, or the risks of sending prompts, files, or secrets to third-party APIs. Because the tool is generic and can target any REST API, the absence of guardrails materially increases the chance an agent will transmit sensitive data or use high-privilege credentials in unsafe contexts.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly instructs the agent to configure bearer-token-authenticated API access and execute arbitrary API operations derived from an OpenAPI spec, but it provides no warning that those commands can send sensitive data to external or internal services or trigger state-changing actions. Because this skill is user-invocable and designed to work against any API profile, it materially increases the risk of unauthorized data exfiltration, destructive API calls, or misuse of privileged credentials if an agent follows the workflow without additional trust checks and confirmation.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.