Agent skill
ultrathink-orchestrator
Auto-invokes ultrathink workflow for any work request (default orchestrator)
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/productivity/ultrathink-orchestrator
SKILL.md
Ultrathink Orchestrator Skill
Purpose
This skill provides automatic orchestration for development and investigation tasks. It detects the task type from keywords and delegates to the appropriate workflow skill (investigation-workflow or default-workflow).
Auto-activation priority is LOW (5) to allow more specific skills to match first. When activated, this orchestrator selects between investigation-workflow and default-workflow based on the user's request keywords.
This skill acts as a thin wrapper around the canonical ultrathink command, following the amplihack pattern of single-source-of-truth for command logic.
Canonical Sources
This skill is a thin wrapper that references canonical sources:
- Primary Command:
.claude/commands/amplihack/ultrathink.md(278 lines) - Workflow Sources:
- Development:
.claude/workflow/DEFAULT_WORKFLOW.md - Investigation:
.claude/workflow/INVESTIGATION_WORKFLOW.md
- Development:
The canonical command contains complete task detection logic, complexity estimation, and orchestration patterns for both investigation and development workflows.
Execution Instructions
When this skill is activated, you MUST:
-
Read the canonical command for task detection logic:
Read(file_path=".claude/commands/amplihack/ultrathink.md")Note: Path is relative to project root. Claude Code resolves this automatically.
-
Detect task type using keywords from the canonical command:
- Investigation keywords: investigate, explain, understand, analyze, research, explore
- Development keywords: implement, build, create, add feature, fix, refactor, deploy
- Hybrid tasks: Both investigation and development keywords present
-
Invoke appropriate workflow skill:
- Investigation:
Skill(skill="investigation-workflow") - Development:
Skill(skill="default-workflow") - Hybrid: Both sequentially (investigation first, then development)
- Investigation:
-
Fallback if skill not found:
- Read workflow markdown files directly
- Investigation:
.claude/workflow/INVESTIGATION_WORKFLOW.md - Development:
.claude/workflow/DEFAULT_WORKFLOW.md - Follow workflow steps as specified
Why This Pattern
Benefits:
- Single source of truth for orchestration logic in canonical command
- No content duplication between command and skill
- Task detection rules defined once, maintained once
- Changes to ultrathink command automatically inherited by skill
Trade-offs:
- Requires Read tool call to fetch canonical logic
- Slight indirection vs. inline implementation
This pattern aligns with amplihack philosophy: ruthless simplicity through elimination of duplication.
Related Files
- Canonical Command:
.claude/commands/amplihack/ultrathink.md - Development Workflow Skill:
.claude/skills/default-workflow/ - Investigation Workflow Skill:
.claude/skills/investigation-workflow/ - Canonical Workflows:
.claude/workflow/DEFAULT_WORKFLOW.md.claude/workflow/INVESTIGATION_WORKFLOW.md
Didn't find tool you were looking for?