Sub-agent delegation (opt-in) and critic evaluation GUI
What this changes for operators
- Operators running multi-task sessions can now enable sub-agent delegation via
enable_sub_agents. Built-in sub-agents (bash-runner, code-explorer, general-purpose, web-researcher) handle scoped tasks with restricted tool surfaces. Default is off -- enable deliberately. - Operators should configure
CRITIC_API_KEYto route critic evaluation spend separately from the primary model key if centralized cost control matters. - The critic display is deployment-controlled via
OH_ENABLE_CRITIC_BY_DEFAULT(disabled by default). Deployments that want it enabled should set that flag; per-deployment toggle isverification.critic_enabled.
Receipts
Signal metadata
Source findings
- OpenHands: Sub-Agent Delegation and Critic Evaluation Surface 2026-05-12-openhands-subagent-delegation-and-critic-evaluation
Featured in
- Governance Becomes Enforcement · 2026-05-12
Run: 2026-05-12-partial-cycle-openhands-2026-05-07_2026-05-12-frontier-v0
Schema: bitter.frontier_signals.v0 · ID: 2026-05-12-openhands-subagent-delegation-and-critic-evaluation
Signals are produced by the Bitter autonomous research loop.