Mission control plugin health

Mission Control Plugin Health

Purpose

Check the health of the Mission Control bridge surfaces that Codex depends on.

The Codex chat agent is not the Mission Control Manager. It is the bridge between the user and the Mission Control Manager.

Use when

• Mission Control seems unreachable or partially broken.
• The user asks whether the plugin or daemon is healthy.
• A setup problem blocks orchestration.

Workflow

1. Call mission_control_plugin_health when available.
2. Review status resources and summarize daemon, MCP, skills, runner registry, runtime folder, local binding, and optional dashboard state.
3. Return a concise health report and likely next fix.

Mission Control calls

Tools:

• mission_control_plugin_health
• mission_control_get_status

Resources:

• mission-control://projects/{project_id}/diagnostics
• mission-control://projects/{project_id}/status

User-facing output

• Check daemon, MCP, skills, Codex CLI, runner registry, runtime folder, localhost binding, and optional dashboard status.
• Report failures as specific components, not a vague 'it is broken.'

Approval behavior

Health reads are read-only. If the user wants a repair action that changes config or restarts components, get explicit approval first.

Never do

• Do not pretend health is good when connectivity is partial.
• Do not dump raw diagnostics logs by default.
• Do not force dashboard UI involvement.

Failure and fallback

If the dedicated health tool is absent, use diagnostics and status resources to produce a best-effort bridge health summary and label missing checks clearly. If the plugin registered with only partial MCP exposure, say that directly instead of flattening it into a generic outage.

Example invocation

Run a Mission Control plugin health check.