Apply suggestion from @npentrel

Author: npentrel

src/oss/deepagents/skills.mdx

@@ -319,10 +319,6 @@ When skills are configured, a "Skills System" section is injected into the agent
 
 Skills can include scripts alongside the `SKILL.md` file (for example, a Python or TypeScript file that performs a search or data transformation). The agent can _read_ these scripts from any backend, but to _execute_ them, the agent needs access to a shell — which only [sandbox backends](/oss/deepagents/sandboxes) provide.
 
-<Warning>
-If your skills include executable scripts, you must use a sandbox backend. Non-sandbox backends (`StateBackend`, `FilesystemBackend`, `StoreBackend`) expose filesystem tools but do not provide an `execute` tool, so the agent cannot run scripts stored there.
-</Warning>
-
 ### Sync skills into the sandbox
 
 When you use a @[CompositeBackend] that routes skills to a @[StoreBackend] (for persistence) while using a sandbox as the default backend, skill files live in the store but the sandbox is where code runs. Use [custom middleware](/oss/langchain/middleware/custom) to upload skill scripts into the sandbox before the agent starts, so the agent can execute them with the `execute` tool.