Publer

• Outcome: Pass
• Latest audit: May 28, 2026, 12:49 PM
• Version:
  1.0.1

Overview

The skill is a disclosed Publer integration for uploading media and publishing or scheduling social posts, with no evidence of hidden behavior or unauthorized data transfer.

Install only if you intend to let an agent use your Publer API key. Review account IDs, media paths, captions, privacy, and schedule before running live post commands, and prefer the dry-run option for anything important or ambiguous.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 92% confidence
• Finding: The skill requires environment secrets and network access to interact with the Publer API, but it does not declare corresponding permissions. This creates a transparency and governance gap: the agent may invoke a networked, credentialed action without explicit permission metadata, increasing the chance of unintended outbound actions or secret-backed operations.

Vague Triggers

Medium

• Confidence: 82% confidence
• Finding: The trigger language is very broad and can match generic requests like 'publish to social media' or 'schedule a post,' which may cause the agent to select this skill in situations the user did not specifically intend to route through Publer. Because the skill performs real network actions using stored API credentials, over-selection increases the risk of unintended posting, scheduling, or data disclosure to a third-party service.

Missing User Warnings

Medium

• Confidence: 73% confidence
• Finding: The upload function sends arbitrary local file contents to a remote third-party API with no built-in confirmation, path restrictions, or safeguards around sensitive files. In an agent context, this increases the chance of accidental exfiltration if an upstream prompt or workflow supplies unintended file paths.

Missing User Warnings

Medium

• Confidence: 75% confidence
• Finding: The post function transmits user-provided text, media references, scheduling data, and account targeting to an external publishing service without any explicit consent or confirmation step. In an agent skill, this can cause unintended external publication or disclosure if invoked with mistaken or manipulated inputs.

VirusTotal

66/66 vendors flagged this skill as clean.