Security audit

Feishu Multi Agent Messaging

Security checks across malware telemetry and agentic risk

Overview

This is a coherent instruction-only Feishu messaging skill, but it handles bot credentials, user IDs, and real chat routing so users should configure it carefully.

Install only if you intend to let OpenClaw send Feishu messages through configured bots. Keep appSecret values out of repositories and shared docs, protect user open_id mapping files, restrict allowlists and bot permissions, verify accountId and target IDs before sending, and avoid posting secrets, personal data, or sensitive document links into group chats.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (4)

Missing User Warnings

Medium

Confidence: 84% confidence
Finding: The README includes concrete bot credential fields (`appId`, `appSecret`), user identifiers (`open_id`, `allowFrom`), and private-message routing examples, but does not warn readers not to commit real secrets or expose user identifiers. In a messaging integration skill, this omission can lead operators to paste production tokens and personal IDs into configs, docs, or repos, causing credential leakage and privacy exposure.

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: The skill explicitly supports relaying messages across multiple Feishu bot accounts and maintaining per-bot user open_id mappings, but it does not warn about privacy, consent, data minimization, or cross-account exposure of message content and identifiers. In a multi-agent messaging context, this omission can lead operators to share user data between bots or group/private contexts without understanding the privacy and compliance implications.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The workflow explicitly requires participants to post status updates, task details, and even document links into a shared Feishu group chat, but it provides no guidance on data classification, least disclosure, or when to move sensitive content to a private channel. In a multi-agent collaboration setting, this increases the chance of leaking internal requirements, user data, links to sensitive documents, or operational details to broader audiences than necessary.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The examples demonstrate direct message sending to users and groups, including @mentions and account impersonation via different bot identities, but provide no guidance on recipient validation, sensitive-content handling, or privacy boundaries. In a multi-agent messaging skill, this omission can lead developers to send task details, links, or user data to the wrong recipient or group, causing unauthorized disclosure or cross-agent data leakage.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.