Wikclawpedia Archive Access

• Outcome: Pass
• Latest audit: May 28, 2026, 4:47 AM
• Version:
  1.0.0

Overview

This is a straightforward Wikclawpedia API wrapper; its main risk is that the submit feature sends user-provided information to an external service for possible publication.

Install this only if you want your agent to read from and submit to Wikclawpedia. Before using submit, review the payload and avoid sending secrets, private/internal data, personal information, or anything you would not want reviewed, stored, attributed, or published.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (3)

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The skill encourages users to submit 'intel' to an external wiki and states submissions may be reviewed and published, but it does not clearly warn that entered data leaves the local environment and could become public. In an agent context, this can lead to accidental disclosure of sensitive prompts, internal URLs, tokens, proprietary data, or user information if an agent forwards data without explicit disclosure and consent controls.

Missing User Warnings

Medium

• Confidence: 83% confidence
• Finding: The submit function transmits caller-supplied intel to an external service via POST with no visible confirmation, warning, or consent checkpoint in this file. In an agent-skill context, this can enable unintended data exfiltration or publication of sensitive user/agent data if higher-level tooling invokes submit on untrusted or sensitive content.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The skill exposes a write-capable `submit` function that sends user-provided intel to an external public service, and the schema explicitly allows a `submitter` name for attribution, but the manifest provides no warning about external transmission, publication, or identity disclosure. This can mislead users or upstream agents into sharing sensitive, proprietary, or identifying data under the assumption that the skill is a neutral local knowledge tool rather than a public submission channel.

VirusTotal

64/64 vendors flagged this skill as clean.