Collaboration Manager

ReviewAudited by ClawScan on May 10, 2026.

Overview

This looks like a legitimate group-chat coordination skill, but it asks to override existing chat rules and its agent identity routing is ambiguous because multiple agents share the same Feishu ID.

Review this skill before installing in a real group chat. It is not obviously malicious, but you should fix the duplicate Feishu agent IDs, make sure it cannot override important AGENTS.md safety rules, and decide how shared task state will be stored and cleared.

Findings (3)

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.

What this means

Agents may respond or stay silent according to this skill even when existing AGENTS.md chat rules say otherwise.

Why it was flagged

The skill explicitly tells the agent to treat its group-chat rules as higher priority than AGENTS.md, which can override existing workspace or group-chat behavior controls.

Skill content
**⚠️ 重要:这个 skill 的规则在群聊中优先于 AGENTS.md 的通用群聊规则。**
Recommendation

Only install if you want this skill to govern group-chat routing, and revise the instructions so they do not override safety, security, or workspace policy rules.

What this means

Messages may be routed to the wrong agent, multiple agents may believe they were addressed, or the intended recipient of a task may be unclear.

Why it was flagged

The artifacts acknowledge that agent identities cannot be distinguished by openId, even though the collaboration logic depends on @ mentions and agent identity.

Skill content
当前配置中所有 agents 使用相同的 openId(`ou_c917d9b06ff30b98c4f45c79219164d5`)... 无法通过 openId 区分 agents
Recommendation

Assign unique Feishu openIds or app identities per agent, and enforce sender/recipient checks before relying on @-based routing.

What this means

Task details and collaboration state may persist in the workspace and influence later agent responses.

Why it was flagged

The skill describes shared workspace task state, which is purpose-aligned but creates persistent context that may be reused by multiple agents.

Skill content
使用共享的任务状态(存储在 workspace)
Recommendation

Define retention, access, review, and cleanup rules for shared task state before using it for sensitive work.