Aoineco Squad Dispatch — Multi-Agent Task Router
ReviewAudited by ClawScan on May 10, 2026.
Overview
The skill’s visible purpose is a task-routing helper, but hidden/obfuscated SKILL.md signals and multi-agent dispatch behavior deserve review before installation.
Before installing, inspect the raw SKILL.md for hidden comments or unusual unicode/control characters. If you use the skill, ask it to show the dispatch plan first, approve execution explicitly, and do not include secrets in tasks that may be routed to other agents.
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.
A user’s agent could potentially see or be influenced by instruction text that is not obvious to the user in normal rendered documentation.
The skill instruction file contains detected unicode-control-character injection indicators and had a hidden comment block removed during neutralization. That kind of hidden or obfuscated instruction surface is not explained by the task-routing purpose and could be used to influence an agent outside the visible documentation.
Pre-scan injection signals: "unicode-control-chars"; "hiddenCommentBlocksRemoved": 1
Review the raw SKILL.md, remove hidden comments and control characters, and only install a version whose instructions are fully visible and explainable.
Task descriptions, priorities, and possibly sensitive work context may be sent to other agents or models when dispatch is used.
The router’s documented workflow includes executing a plan through spawned agent sessions. That is purpose-aligned for a multi-agent dispatcher, but it means task content may be delegated across agent/session boundaries.
dispatcher.dispatch(plan) — Execute (via sessions_spawn)
Use the planning output for review before dispatch, require user confirmation for execution, and avoid placing secrets or private data in task descriptions unless the receiving agents are trusted.