dipeo-package-maintainer
Router skill for DiPeO runtime Python code (execution handlers, service architecture, domain models, LLM infrastructure). Use when task mentions node handlers, EventBus, ServiceRegistry, Envelope pattern, or domain logic. For simple tasks, handle directly; for complex work, escalate to dipeo-package-maintainer agent.
When & Why to Use This Skill
This Claude skill serves as an intelligent router and maintenance tool for DiPeO runtime Python codebases. It streamlines development by managing service architecture, domain models, and LLM infrastructure, providing a bridge between simple code lookups and complex agent-led refactoring. By automating the decision-making process for task escalation, it ensures that Python node handlers, EventBus integrations, and ServiceRegistry configurations are handled by the most appropriate resource.
Use Cases
- Architectural Pattern Lookup: Quickly identify and retrieve implementation examples for the Envelope pattern, EventBus usage, or ServiceRegistry configurations within the /dipeo/ directory.
- Minor Code Refactoring: Perform low-complexity updates, such as fixing typos, minor handler tweaks, or small refactors involving fewer than 20 lines of code across limited files.
- Intelligent Task Escalation: Automatically evaluate the complexity of incoming tasks to determine if they can be handled directly or require escalation to the dipeo-package-maintainer agent.
- Domain Logic Exploration: Facilitate read-only reviews and understanding of complex domain models, conversation management strategies, and memory implementations.
- Workflow Optimization: Coordinate between different specialized agents (like codegen-pipeline or backend agents) when tasks involve generated code issues or CLI configurations.
| name | dipeo-package-maintainer |
|---|---|
| description | Router skill for DiPeO runtime Python code (execution handlers, service architecture, domain models, LLM infrastructure). Use when task mentions node handlers, EventBus, ServiceRegistry, Envelope pattern, or domain logic. For simple tasks, handle directly; for complex work, escalate to dipeo-package-maintainer agent. |
| allowed-tools | Read, Grep, Glob, Bash, Skill |
DiPeO Package Maintainer Router
Domain: Runtime Python in /dipeo/ (application, domain, infrastructure - EXCLUDING codegen and generated code).
Quick Decision: Skill or Agent?
✅ Handle Directly (This Skill)
- Simple changes: <20 lines, 1-2 files
- Read-only tasks: Understanding patterns, reviewing implementation
- Pattern lookup: Envelope usage, service registration examples
- Small fixes: Typos, minor refactors, simple handler tweaks
❌ Escalate to Agent
- New node handlers: Creating handlers with complex logic
- Service architecture: EventBus integration, ServiceRegistry configuration
- Domain model changes: Conversation management, memory strategies
- Multi-file refactoring: Affects >2 files or cross-layer changes
- Generated code issues: If generated APIs don't work, escalate to dipeo-codegen-pipeline first
Agent: Task(dipeo-package-maintainer, "your detailed task description")
Documentation Sections (Load On-Demand)
Use Skill(doc-lookup) with these anchors when you need detailed context:
Architecture & Ownership:
docs/agents/package-maintainer.md#your-domain-of-expertise- Code layers and ownershipdocs/agents/package-maintainer.md#core-architectural-principles- Service architecture, LLM integration
Implementation:
docs/agents/package-maintainer.md#your-responsibilities- Node handlers, services, GraphQL resolvers
Patterns & Imports:
docs/agents/package-maintainer.md#common-patterns- Envelope, service registry, node handlersdocs/agents/package-maintainer.md#key-import-paths- Common imports
Escalation:
docs/agents/package-maintainer.md#when-you-need-help-escalation-paths- When to escalate
Example doc-lookup call:
python .claude/skills/doc-lookup/scripts/section_search.py \
--query "envelope-pattern" \
--paths docs/agents/package-maintainer.md \
--top 1
Escalation to Other Agents
To dipeo-codegen-pipeline: Generated code issues, TypeScript specs, new node types, GraphQL schema To dipeo-backend: CLI commands, server config, database schema, MCP server
Typical Workflow
- Assess complexity: Simple pattern lookup vs. complex implementation
- If simple: Load relevant section via
Skill(doc-lookup), provide guidance or make small change - If complex: Escalate with
Task(dipeo-package-maintainer, "task details") - Generated code problems: First check if it's a generation issue → escalate to dipeo-codegen-pipeline