From space-agents
HOUSTON spawns Pathfinder, Builder, Inspector per task. Best for medium features (4-10 tasks).
npx claudepluginhub thebrownproject/space-agents --plugin space-agentsThis skill uses the workspace's default tool permissions.
HOUSTON coordinates execution by spawning fresh agents for each task. Prevents context rot.
Creates isolated Git worktrees for feature branches with prioritized directory selection, gitignore safety checks, auto project setup for Node/Python/Rust/Go, and baseline verification.
Executes implementation plans in current session by dispatching fresh subagents per independent task, with two-stage reviews: spec compliance then code quality.
Dispatches parallel agents to independently tackle 2+ tasks like separate test failures or subsystems without shared state or dependencies.
HOUSTON coordinates execution by spawning fresh agents for each task. Prevents context rot.
bd show FEATURE_ID to get detailsbd update FEATURE_ID --status in_progressbd list --parent FEATURE_ID --status open (pick highest priority)bd update TASK_ID --status in_progressbd close TASK_IDbd close FEATURE_IDUse Task tool with run_in_background: false (wait for completion).
CRITICAL: Always pass task_id and feature_id explicitly. Agents will run bd show to fetch authoritative details from Beads.
Pathfinder (subagent_type: "space-agents:mission-pathfinder"):
"Explore codebase for task [TASK_ID] under feature [FEATURE_ID].
Run `bd show [TASK_ID]` and `bd show [FEATURE_ID]` first to get full context.
Add findings as a [PATHFINDER] comment on the task bead."
Builder (subagent_type: "space-agents:mission-builder"):
"Execute task [TASK_ID] for feature [FEATURE_ID].
Run `bd show [TASK_ID]` and `bd show [FEATURE_ID]` first to get full context.
Read [PATHFINDER] comment for codebase findings.
Use **Tests:** checklist in task description as TDD targets."
Inspector (subagent_type: "space-agents:mission-inspector"):
"Review task [TASK_ID] for feature [FEATURE_ID].
Run `bd show [TASK_ID]` and `bd show [FEATURE_ID]` first to get requirements.
Verify each **Tests:** item in task description. Report pass/fail count."
After Inspector returns, HOUSTON decides:
| Result | Action |
|---|---|
| All pass | Mark complete, continue to next task |
| Minor issues | Log warning, continue |
| Blocker found | Create bug via bd create -t bug, ask user |
| Critical issue | Halt, escalate to user |
HOUSTON: Feature complete. {N} tasks executed via orchestrated mode.
Pathfinder/Builder/Inspector cycle completed for each task.