Feishu Thread Forward

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:15 PM UTC
• Version:
  1.0.0

Overview

This skill does what it claims: it forwards Feishu threads using configured Feishu bot credentials, but users must verify the destination because forwarding can reveal thread context.

Install only if you want the agent to forward Feishu topics using your configured Feishu bot. Before each use, confirm the source thread, recipient ID, recipient type, group sensitivity, and whether the thread is appropriate to share. Keep the Feishu app secret and tenant tokens out of chat, logs, and screenshots, and grant the bot only the permissions it needs.

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: 91% confidence
• Finding: This skill forwards an entire Feishu thread/topic to another recipient, which can disclose prior conversation context to a new user, group, or topic. Without an explicit user-facing warning, an operator may trigger sharing without appreciating that this is cross-chat data disclosure, increasing the chance of accidental information leakage.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The skill documents reading app credentials from a local secrets file but gives no warning about secret handling, least privilege, or avoiding exposure in logs and outputs. Guidance that normalizes direct credential access without safeguards increases the risk of token theft, accidental disclosure, and broader compromise of Feishu resources accessible to the bot.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The script will forward a Feishu thread to any recipient ID supplied on the command line without an interactive confirmation, allowlist, or policy check. In an agent context, this increases the risk of unintended or socially engineered exfiltration of internal discussion content to the wrong user, group, or thread.

VirusTotal

64/64 vendors flagged this skill as clean.