Feishu Multi Agent Messaging

Security checks across malware telemetry and agentic risk

Overview

This is a coherent instruction-only Feishu messaging guide; it does involve bot credentials, user ID mappings, and external message sending, but these are disclosed and aligned with its purpose.

This skill appears safe to install as an instruction-only guide, but only use it if you are comfortable granting OpenClaw Feishu bot messaging authority. Protect Feishu app secrets, keep user ID mappings private, restrict allowed senders/groups, and verify accountId and target IDs before sending real messages.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Risk analysis

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

ASI03: Identity and Privilege Abuse

Low

What this means

Anyone with these bot secrets or misconfigured account bindings could send Feishu messages as those bots.

Why it was flagged

The skill instructs users to configure multiple Feishu bot credentials and choose which bot identity sends a message via accountId.

Skill content

"accounts": { "default": { "appId": "cli_xxx", "appSecret": "xxx" }, "coder": { "appId": "cli_yyy", "appSecret": "yyy" } }

Recommendation

Use least-privilege Feishu apps, keep appSecret values out of shared files and version control, restrict allowed users/groups, and rotate secrets if exposed.

ASI02: Tool Misuse and Exploitation

Low

What this means

The agent can send messages to Feishu users or groups if the environment grants it access to the message tool and bot credentials.

Why it was flagged

The skill documents direct use of the message tool to send Feishu DMs or group messages to specified targets.

Skill content

message action=send \
  channel=feishu \
  accountId="coder" \
  target="ou_user_openid_seen_by_coder_bot" \
  message="码农收到任务，开始执行"

Recommendation

Confirm the intended accountId, target, and message content before sending, especially for group chats or messages that notify real users.

ASI06: Memory and Context Poisoning

Low

What this means

Incorrect, stale, or exposed mappings could reveal user identifiers or cause messages to be sent to the wrong person.

Why it was flagged

The skill recommends persisting a cross-bot user open_id mapping table, which may contain privacy-sensitive identifiers and influences future message routing.

Skill content

**推荐位置：** `SESSION-STATE.md` 或独立的配置文件

Recommendation

Store mappings in a protected location, review changes, avoid unnecessary identifiers, and verify target IDs before sending sensitive messages.

ASI07: Insecure Inter-Agent Communication

Low

What this means

Misconfigured bindings or targets could route a task or message through the wrong bot or to the wrong Feishu chat.

Why it was flagged

The documented workflow routes messages between multiple agents and Feishu bot accounts through the Gateway, so identity and routing boundaries depend on correct configuration.

Skill content

OpenClaw Gateway ... Message Router ... - accountId 匹配 ... - target 解析 ... - 消息路由

Recommendation

Keep agent-to-account bindings explicit, test with non-sensitive messages first, and maintain allowlists for users and groups.