Smart Spawn
PassAudited by ClawScan on May 10, 2026.
Overview
This instruction-only skill is coherent with its stated purpose, but it sends task descriptions to a third-party API and may use that API's output to spawn model sessions.
This skill appears benign and instruction-only. Before using it, remember that your task description is sent to ss.deeflect.com and that its recommendations can drive model selection or parallel spawning. Use explicit budgets, avoid sharing sensitive task details, and review swarm/decomposition plans before letting the agent spawn many sessions.
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.
The API can influence which model is used and may affect cost, performance, or the number of spawned sessions.
The skill chains a third-party API recommendation into the model used for spawning a session. This is expected for the skill, but users should ensure model choice, budget, and spawning behavior remain under their control.
Use the returned `id` as the `model` parameter in `sessions_spawn`
Review model recommendations before spawning expensive or parallel sessions, and set explicit budget/count limits when using the skill.
Any details included in the task description may be shared with ss.deeflect.com.
Task descriptions are sent to an external Smart Spawn API. This is disclosed and necessary for the recommendation service, but task text may contain sensitive project or business details.
GET ss.deeflect.com/api/pick?task=<description>&budget=<tier>
Avoid sending secrets, private customer data, or confidential plans in task descriptions unless you trust the API provider and its data handling practices.
A complex task could result in multiple spawned sessions, increasing spend or spreading an incorrect plan across subtasks.
The skill supports decomposition and parallel swarm workflows based on external API output. This is part of the advertised functionality, but unbounded fan-out could increase cost or propagate a poor decomposition across multiple spawned tasks.
use `/api/decompose` or `/api/swarm` to break it into subtasks, spawn each with its recommended model
Approve decomposition plans before execution and cap the number of spawned sessions for complex or high-cost work.