Stats
Actions
Tags
Help us improve
Share bugs, ideas, or general feedback.
From hope
Orchestrates the full hope pipeline end-to-end for a task: session setup, intent clarification, approach shaping, optional consult/bond, and execution.
npx claudepluginhub saadshahd/moo.md --plugin hopeHow this command is triggered — by the user, by Claude, or both
Slash command
/hope:fullThe summary Claude sees in its command listing — used to decide when to auto-load this command
# /hope:full ORCHESTRATE. Run the hope pipeline end-to-end. **Task:** $0 ## Session Setup Ask: "How would you like to work? Autonomous / Collaborative [default] / Guided?" Detect session type from the task: Build, Debug, Plan, or Reflect. Emit: `[SESSION] Pipeline: [phases] | Engagement: [level]` ## Pipeline 1. **Intent** — Clarify WHAT. If the task arrives with structured input (proposal, specs, task list), validate rather than re-clarify. Complete when you have a clear objective and acceptance criteria. 2. **Shape** — Decide HOW. Get expert perspectives on the approach. Skip for ...
/planCaptures user intent via 5 structured questions, creates strategic execution plan, saves to .claude/session-plan.md and session-intent.md for review.
/create-planCreates an implementation plan from arguments using the planning skill. Requires an active session started with /session-start or /session-continue.
/plan-modeCreates multi-agent task hierarchy for planning consultation on a feature, dispatching specialist consultations in parallel and synthesizing an implementation plan (planning only).
/plannewCreates new development plan from scratch using feature/project description. Enters plan mode, enforces TDD/DRY/SOLID principles, sets up tasks with strict rules.
/create-planCreates hierarchical project plans for solo agentic development, generating briefs, roadmaps, and phase plans from a description of what to plan.
/create-planCreates detailed implementation plans through interactive research and iteration for tasks or files, with configurable autonomy: autopilot, critical, verbose.
Share bugs, ideas, or general feedback.
ORCHESTRATE. Run the hope pipeline end-to-end.
Task: $0
Ask: "How would you like to work? Autonomous / Collaborative [default] / Guided?"
Detect session type from the task: Build, Debug, Plan, or Reflect. Emit:
[SESSION] Pipeline: [phases] | Engagement: [level]
Intent — Clarify WHAT. If the task arrives with structured input (proposal, specs, task list), validate rather than re-clarify. Complete when you have a clear objective and acceptance criteria.
Shape — Decide HOW. Get expert perspectives on the approach. Skip for Reflect sessions or trivially clear tasks. Complete when you have a concrete approach and first step.
Consult (optional) — Shape already includes consultation. This is an additional review pass if the user wants a second opinion. For Reflect sessions, consult is the primary stage.
Bond (when needed) — If the task spans multiple modules and warrants parallel agents, design and create a team.
Execute — For solo paths, proceed with the shaped approach.
The pipeline is a map, not a rail. Skip stages that don't apply. Come back to earlier stages if the task reveals new information.