goal-plan — Claude Code skill: action planning with preconditions
A skill that builds step-by-step plans for complex tasks, tracks what conditions must be met before each step, estimates costs, and switches to an alternative path if something fails.
npx skills add ruvnet/ruflo --skill goal-plan -a claude-codeWhat goal-plan does
goal-plan uses Goal-Oriented Action Planning (GOAP) — a search algorithm that finds the optimal sequence of actions through state space. The skill takes a goal description, identifies what must change from the current state to the desired state, and builds a plan from available actions accounting for preconditions, effects, and cost of each step.
Internally it uses task management tools (task_create, task_list, task_status), memory search (memory_search, memory_store), workflow execution (workflow_execute), and trajectory hooks (trajectory-start, trajectory-step, trajectory-end) to track and execute the plan. Actions run in dependency order: preconditions are verified before each action, effects verified after.
If an action fails or new information appears, the skill recalculates the path from the new state, updates remaining tasks, and continues. Successful plans are stored in memory under the namespace goap-plans for reuse.
What it helps with
- split a complex goal into ordered actions with precondition checking
- select the lowest-cost path among multiple action sequences
- replan automatically when conditions change or an action fails
- record the solution trajectory and store the successful plan for later use
Example prompts for the agent
- Migrate a React project from JavaScript to TypeScript — find all files, determine the change order accounting for dependencies
- A production microservice is crashing — figure out root causes, determine the check order, switch to a backup if a path is blocked
- Design a database schema for an app — identify which migrations are needed and in what sequence to deploy them
How to install goal-plan
1Skills CLI (Vercel)
npx skills add ruvnet/ruflo --skill goal-plan -a claude-code2Manually: put the files in place
~/.claude/skills/goal-plan/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/goal-plan/
1Skills CLI (Vercel)
npx skills add ruvnet/ruflo --skill goal-plan -a codex2Manually: put the files in place
.agents/skills/goal-plan/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add ruvnet/ruflo --skill goal-plan -a cursor2Manually: put the files in place
.cursor/skills/goal-plan/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install ruvnet/ruflo goal-plan2Manually: put the files in place
.github/skills/goal-plan/SKILL.md1Skills CLI (Vercel)
npx skills add ruvnet/ruflo --skill goal-plan -a gemini-cli2Platform CLI
gemini skills install https://github.com/ruvnet/ruflo --path plugins/ruflo-goals/skills/goal-planManually: put the files in place
.devin/skills/goal-plan/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add ruvnet/ruflo --skill goal-plan -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
From the same set104
The whole set →Similar
Questions about goal-plan
How do I install goal-plan in Claude Code?
Run npx skills add ruvnet/ruflo --skill goal-plan -a claude-code. The skill will appear in ~/.claude/skills/goal-plan/SKILL.md.
What does goal-plan do?
The skill takes a goal description and current state, analyzes available actions with their preconditions and effects, builds an optimal plan using A* search, executes it step by step, and switches to an alternative route if conditions change.
When does the skill trigger replanning?
The skill switches to a new plan if an action fails, unexpected effects are detected, the goal definition changes, estimated cost is exceeded, or an external dependency becomes unavailable.
Source text
SKILL.md · MIT
--- name: goal-plan description: Create and execute Goal-Oriented Action Plans (GOAP) with precondition analysis, cost optimization, and adaptive replanning argument-hint: "<goal-description>" allowed-tools: mcp__plugin_ruflo-core_ruflo__task_create mcp__plugin_ruflo-core_ruflo__task_list mcp__plugin_ruflo-core_ruflo__task_status mcp__plugin_ruflo-core_ruflo__task_assign mcp__plugin_ruflo-core_ruflo__task_update mcp__plugin_ruflo-core_ruflo__task_complete mcp__plugin_ruflo-core_ruflo__task_summary mcp__plugin_ruflo-core_ruflo__memory_store mcp__plugin_ruflo-core_ruflo__memory_search mcp__plugin_ruflo-core_ruflo__neural_predict mcp__plugin_ruflo-core_ruflo__workflow_create mcp__plugin_ruflo-core_ruflo__workflow_execute mcp__plugin_ruflo-core_ruflo__workflow_status mcp__plugin_ruflo-core_ruflo__hooks_intelligence_trajectory-start mcp__plugin_ruflo-core_ruflo__hooks_intelligence_trajectory-step mcp__plugin_ruflo-core_ruflo__hooks_intelligence_trajectory-end Bash Read Write Edit --- # Goal Plan Create and execute intelligent plans using Goal-Oriented Action Planning (GOAP). ## When to use When you have a complex objective that requires multiple steps, has dependencies between steps, and may need adaptive replanning as conditions change. ## Steps 1. **Define goal state** — what does "done" look like? List concrete success criteria 2. **Assess current state** — what's true now? What assets, code, infrastructure exist? 3. **Identify gap** — what must change between current and goal state? 4. **Inventory actions** — list available actions with: - Preconditions (what must be true before this action) - Effects (what becomes true after this action) - Cost estimate (time, complexity, risk) 5. **Generate plan** — find the optimal action sequence using A* through the state space 6. **Record trajectory** — call `mcp__plugin_ruflo-core_ruflo__hooks_intelligence_trajectory-start` to begin tracking 7. **Create tasks** — call `mcp__plugin_ruflo-core_ruflo__task_create` for each action in the plan 8. **Execute** — work through tasks in dependency order: - Before each action: verify preconditions still hold - After each action: verify effects achieved - Record each step via `mcp__plugin_ruflo-core_ruflo__hooks_intelligence_trajectory-step` 9. **Monitor & replan** — if an action fails or produces unexpected results: - Reassess current state - Recalculate optimal path from new state - Update remaining tasks 10. **Complete trajectory** — call `mcp__plugin_ruflo-core_ruflo__hooks_intelligence_trajectory-end` 11. **Store successful plan** — call `mcp__plugin_ruflo-core_ruflo__memory_store` with namespace `goap-plans` ## Plan output format ``` Goal: [concrete objective] Current State: [key facts] Plan Cost: [estimated effort] Steps: 1. [action] — precondition: [X], effect: [Y], cost: [Z] 2. [action] — precondition: [Y], effect: [W], cost: [Z] ... Risk Factors: [what could force a replan] Fallback: [alternative approach if primary path fails] ``` ## Replanning triggers - Action fails (precondition no longer met) - Unexpected side effects detected - New information changes goal definition - Cost exceeds threshold - External dependency becomes unavailable
© ruvnet · MIT · ruvnet/ruflo