dsh-plugin-subagent-roles

LIMITATIONS

Known limitations

- Tools registered into a child's own scope are not affected by a role's tool policy; the core applies restrictions to inherited tools only. The delegation runtime and some tool plugins register per agent, so a child can end up with a small number of tools beyond its allow list. - Hiding a tool removes its schema and any scope-aware prompt guidance. Prompt sections with static text stay in the child's prompt — which is what `childPromptTrim` removes, and why `tools` mode needs no maintenance while `full` mode matches section names that a future core release may rename (a renamed section stops being trimmed; nothing breaks). - A role persona replaces the deployment persona prefix for the child. The persona suffix, such as the working-directory line, is kept. - `respectModelSelection` prefers the policy a Session captured (the same durable projection the official delegation tool writes) and falls back to the live `subagent-model-selection` setting, which is what seeds a fresh Session. A change therefore applies to Sessions that have not captured a policy yet. - The plugin contributes one card (Settings → Plugins → Plugin configuration) and no Settings navigation group of its own. A card is the host convention for plugin configuration; a dedicated section like the plugin market's exists because that plugin owns a whole browsing page. - `childPromptTrim: 'off'` in the row config unmounts the Settings card with the trim: no namespace is registered, so the Plugins tab has no key to dispatch (deliberate — an operator's `off` is a kill switch the UI must not be able to flip back). The tool-call budget rides that same namespace, so `off` also removes the budget knobs from Settings; budgets then come from role files and the row config only. To stop trimming while keeping the card, set the CARD's mode to `off`, or use `tools` in the row config. - The Settings card covers the default row's namespace (`subagent-roles`). A second row renamed through `toolName` owns a different namespace and therefore shows no card. - `childPromptTrim` applies to EVERY subagent assembly the host plane observes, not only to children this plugin started: the runtime records no role marker on a child, so `subagent` and counterpart rows benefit too. Tools registered in an agent's own scope (`subagent`, `list_agents`) are absent from the registry's global view, so their guidance is never trimmed — a child that really can call one keeps its instructions. - The delegation tool declares no `timeoutMs` unless you set one; an unbounded foreground delegation can outlive the conversation that started it. Bound long runs with `timeoutMs`, `maxDepth`, or the dispatch prompt. - The tool-call budget is enforced **reactively**: the call that crosses the limit is already in the session log, so the child is stopped before that call finishes rather than being prevented from starting it. The count itself is exact. - A remote transport that publishes no local child agent (`SubagentRun.localAgent` is undefined, as with an out-of-process provider) runs its child outside this process, so its `tool/call` events never reach the guard: the budget can neither count nor stop that child, and the notice cannot be delivered. The plugin warns at delegation time rather than reporting a role as guarded when it is not. In-process transports are unaffected. - `timeoutMs` and the budget are independent, and so are the ways they surface — but if a budget stop has already begun and the tool-call deadline then expires while the stopped run is still being torn down, the core's own timeout wrapper replaces the result with its `TOOL_TIMEOUT` error. The child is stopped either way; only the report can be the later guard's. - `scope: session` enforcement is gated to once per turn: an over-budget continuable child that a parent wakes again is stopped again, but the calls a stopped batch never started do not each repeat the stop. - Conversely, per-turn gating means a `scope: delegation` budget that has alr