advise says a subagent cannot start another subagent, which is wrong #28

Found while reviewing #26 for @GauranshMathur, who asked for it to be fixed.

What is wrong

README.md and skills/advise/SKILL.md say "a subagent cannot start another subagent", and so tell the skill to always skip the Fable advisor inside a subagent and run Codex alone.

The Claude Code docs say otherwise (https://code.claude.com/docs/en/sub-agents.md, "Let subagents spawn their own subagents"): "By default, a subagent can spawn subagents of its own, up to three layers below the main conversation." The Agent tool is withheld only at the depth limit, or when the subagent's definition leaves it out. AskUserQuestion is removed from every subagent, so that half of the sentence is right.

Decisions made

• Inside a subagent, the skill dispatches the Fable advisor as usual when it has the Agent tool. It runs Codex alone, and reports Fable as skipped, only when it does not.
• The rule against AskUserQuestion inside a subagent stays, and so does handing a disagreement back to the parent.
• The README wording is corrected to match.