Feishu Parallel Dispatch
PassAudited by ClawScan on May 3, 2026.
Overview
This instruction-only Feishu helper is coherent and disclosed, but users should expect substantial requests to be delegated to background child agents that may report back later.
This skill appears safe for its stated purpose, but install it only if you want substantial Feishu requests to run asynchronously in child sessions. For sensitive or high-impact tasks, clearly state scope, privacy expectations, and whether you want synchronous handling instead.
Findings (2)
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.
Substantial Feishu requests may keep running in separate child sessions and later post results or progress without the main chat staying occupied.
This shows the skill intentionally creates background child tasks that continue after the foreground reply. That is purpose-aligned and disclosed, but users should understand the asynchronous behavior.
For any substantial Feishu request, prefer: ... immediate `sessions_spawn` ... end the current turn ... let the child task announce its own result later
Install only if you want this asynchronous dispatch behavior; explicitly ask for a synchronous answer when needed and use task status checks for visibility.
Sensitive details included in a Feishu request could be delegated to a child agent context selected by the assistant.
The skill routes work to different child-agent contexts. This is expected for a dispatch skill, but it means user prompt content may be shared with the chosen child agent.
Pick the child agent with the clearest fit: - `research` ... - `office` ... - `family` ... - current agent
Avoid including unnecessary sensitive information, and specify the intended agent or privacy boundary when a task involves personal, family, office, or other compartmentalized content.