mission-opencode

To delegate low-cost TDD and debugging work to an OpenCode CLI agent, the mission orchestrator runs OpenCode synchronously, checks availability, selects the configured model, and parses the output to determine if the task completed successfully or requires escalation.

Mission orchestration for coding and testing tasks is a process where an orchestrator delegates low-cost work to external agents like OpenCode. It manages task execution, parses completion status, and emits structured mission-reports to decide whether to escalate to a primary sub-skill.

You can check if OpenCode is available by running a pre-flight usability check using the command `agents check opencode`. This verifies that the OpenCode CLI agent is usable before attempting synchronous invocation for debugging or TDD work.

When an OpenCode task is blocked, the orchestrator captures stdout and stderr, parses for a blocked status, and emits a structured mission-report block. This report includes the exit code, runtime, and diff summary, allowing the orchestrator to escalate to the primary sub-skill.

No, mission-opcode requires a mission orchestrator to function. It is designed as a delegation wrapper that executes OpenCode synchronously when the orchestrator wants to run `mission-tdd` or `mission-debug` at a cheap cost tier, relying on the orchestrator for configuration and escalation.

Limitations include requiring a successful pre-flight usability check and fitting within the cheap cost tier. If OpenCode is unavailable or returns a blocked status, the orchestrator must escalate the debug work to the primary sub-skill instead of retrying the external agent.