con un clic
sub-agent-patterns
Load when operating as a sub-agent invoked via the agent tool — covers handoff mode detection, autonomous execution, clarification protocol, and the minimum completion report contract.
Menú
Load when operating as a sub-agent invoked via the agent tool — covers handoff mode detection, autonomous execution, clarification protocol, and the minimum completion report contract.
Load before planning or routing any task. Required by mission-control — maps task archetypes to ordered sequences of specialist agents, with handoff context and success criteria for each step.
Load before reviewing any phase. Required by all internal reviewer agents — covers adversarial review philosophy, the approve/reject verdict contract, and escalation policy.
Load before running a comprehensive review. Covers scope resolution, pre-flight tool checks, parallel specialist dispatch, synthesis, and output format.
Recipe reference for installing tools in the container environment. Used by bootstrap and specialist reviewer agents to look up the correct install command for any tool.
Load before any API or CLI interface review. Required by the api-reviewer agent — covers REST naming, HTTP semantics, error consistency, breaking changes, CLI conventions, pagination, auth, and a three-tier severity model.
Load before any code review. Required by the reviewer agent — covers what to examine, smell categories, complexity thresholds, dead code signals, and a three-tier severity model.
| name | sub-agent-patterns |
| description | Load when operating as a sub-agent invoked via the agent tool — covers handoff mode detection, autonomous execution, clarification protocol, and the minimum completion report contract. |
| license | AGPL-3.0-or-later |
| allowed-tools | read |
Reference patterns for agents invoked by another agent via the agent tool. These patterns apply only in handoff mode; interactive sessions still use normal user-facing behaviour.
What it is: Handoff mode describes an agent-to-agent invocation made through the agent tool rather than a human starting an interactive session.
How it is recognised:
mission-controlKey notes:
In handoff mode, the invocation serves as approval to execute the full assigned task.
Key notes:
Handoff mode keeps clarification inside the agent-to-agent exchange rather than turning it into a direct user question.
Key notes:
CLARIFICATION_NEEDED tag identifies issues the caller must resolve.Example:
CLARIFICATION_NEEDED: acceptance criterion 3 is contradictory — does "immediate" mean synchronous or within one second?
Every handoff-mode execution ends with a structured report that gives the caller enough information to continue orchestration.
Required fields:
completed or blockednone, or each blocker with a CLARIFICATION_NEEDED tag or exact errorKey notes: