ai-pr-analysis-companion

• Outcome: Pass
• Latest audit: May 28, 2026, 8:43 PM
• Version:
  1.0.0

Overview

This skill does what it says: it reviews GitHub pull requests and publishes a structured report to Feishu, with no hidden executable code found.

Before installing, make sure your team is comfortable sending PR-derived analysis to Feishu. For private or security-sensitive repositories, confirm the Feishu document destination and access controls, and avoid publishing reports that contain secrets or proprietary details beyond the intended audience.

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 (4)

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The README explicitly states that PR review results will be published to Feishu, an external collaboration platform, but does not warn users that code-derived content, review comments, or potentially sensitive repository metadata may leave the original GitHub context. In a PR-analysis skill, this increases the risk of unintentional data disclosure, especially for private repositories, security findings, or proprietary code details.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The skill explicitly states that it will fetch PR diffs and comment history, then generate and publish a report to Feishu, but it does not warn that potentially sensitive repository code and discussion data will be transmitted to external systems. This can lead users to expose proprietary code, internal discussions, or security-relevant review context without informed consent.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The documented workflow retrieves full PR diffs and historical comments, which may contain secrets, proprietary code, incident context, or reviewer-only discussion, yet no privacy or sensitivity warning is provided. In a PR-analysis skill, this omission is more dangerous because the normal operation encourages bulk collection of high-value engineering data.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The report publication step sends generated review output to Feishu but does not disclose that the report may be visible to broader collaborators or contain sensitive excerpts derived from the PR. This creates a real risk of unintended internal data exposure, especially if access controls on the destination document are broad or misconfigured.

VirusTotal

65/65 vendors flagged this skill as clean.