Sdd Pipeline References
Defines agent-flow rules, file-content schemas, and shared prompt conventions for SDD pipeline agents (specifier, planner, builder, verifier, committer) within the agents-uc-taskmanager suite.
What Is Sdd Pipeline References?
Sdd Pipeline References is a technical resource designed to provide well-defined rules and schemas for agents operating within a software development pipeline. Specifically, it targets agents such as specifier, planner, scheduler, builder, verifier, and committer. Rather than being invoked directly by end users, these references serve as internal documentation and contract definitions that facilitate consistent, predictable, and interoperable agent behavior across the pipeline.
The Sdd Pipeline References skill is part of the larger "agents-uc-taskmanager" suite, and its source is available on GitHub. It aggregates multiple markdown documents that define process flows, file content schemas, shared prompt conventions, and more, all of which are critical for orchestrating automated workflows in complex development environments.
Why Use Sdd Pipeline References?
In modern software engineering, pipelines are often composed of modular and autonomous agents, each responsible for specific tasks. As pipelines grow in complexity, ensuring that each agent interacts with others in a predictable and standardized way becomes a significant challenge. Misalignment in expectations, data formats, or process flows can lead to errors, maintenance headaches, and slower delivery cycles.
Sdd Pipeline References addresses these challenges by:
- Standardizing Communication: It defines the structure of inputs, outputs, and intermediate artifacts, ensuring that agents understand how to process and generate data.
- Reducing Integration Overhead: With clear schemas and rules, developers can integrate new or updated agents into the pipeline more quickly and with fewer errors.
- Enhancing Maintainability: When each agent follows the same conventions, updating the pipeline or troubleshooting issues becomes much more manageable.
- Promoting Scalability: As teams grow and pipelines become more complex, the ability to extend or swap out agents without breaking the workflow is critical. Sdd Pipeline References provides the contract that enables this flexibility.
How to Get Started
To begin using Sdd Pipeline References in your own agent-based pipeline:
-
Clone the Repository:
Retrieve the reference materials from GitHub:git clone https://github.com/davepoon/buildwithclaude.git cd buildwithclaude/plugins/agents-uc-taskmanager/skills/sdd-pipeline -
Review the Reference Documents:
Thereferencesdirectory contains markdown files, each documenting a specific aspect of pipeline behavior:agent-flow.md: Defines the overall agent flow.file-content-schema.md: Outlines file format expectations.shared-prompt-sections.md: Describes shared prompt structures.context-policy.md: Provides sliding window rules for context management.xml-schema.md: Specifies XML communication formats.work-activity-log.md: Details activity log conventions.
-
Integrate Into Your Agents:
Agents should be implemented to consume and produce data according to these references. For example, when writing a builder agent, ensure its output matches the format described infile-content-schema.md. -
Validate Agent Conformance:
Use the schemas and rules as validation checkpoints during development and testing. This can be enforced programmatically or via code review.
Key Features
Sdd Pipeline References delivers several essential features:
-
Comprehensive Flow Definitions:
agent-flow.mdprovides a step-wise breakdown of the agent pipeline, clarifying agent responsibilities and transitions. -
Explicit File Content Schemas:
file-content-schema.mdspecifies the exact JSON or Markdown structures agents must emit and consume. This is particularly useful for tasks such as task specification, planning, and result reporting.Example:
{ "task_id": "12345", "description": "Implement authentication flow", "plan": { "steps": [ "Design data model", "Implement login endpoint", "Write tests" ] } } -
Shared Prompt Section Rules:
shared-prompt-sections.mdstandardizes prompt fields such as TASK ID, PLAN.md's seven fields, and WORK-LIST, ensuring agents share a unified vocabulary. -
Context Management Policies:
context-policy.mdoutlines rules for context retention and sliding window mechanisms, crucial for agents operating on large and evolving state spaces. -
XML Communication Schema:
xml-schema.mddefines XML formats for agent-to-agent or agent-to-service message passing.Example:
<Task> <Id>12345</Id> <Description>Implement authentication</Description> <Status>In Progress</Status> </Task> -
Work Activity Log Standards:
work-activity-log.mdensures activity logs are consistently structured for auditing, debugging, and traceability.
Best Practices
-
Strict Schema Adherence:
Always validate that agent inputs and outputs match the schemas defined in the references. Automated schema validation tools can be integrated into CI/CD pipelines. -
Version Control:
Treat the references as versioned contract documents. If you need to update a schema, communicate clearly and coordinate agent updates accordingly. -
Documentation:
Encourage agent developers to cite specific reference files when implementing or updating agents. This increases transparency and reduces onboarding time for new contributors. -
Testing:
Write unit and integration tests that verify agent compliance with the reference schemas and flow rules. -
Separation of Concerns:
Avoid embedding reference logic directly in agent code. Instead, use adapters or interfaces that can be updated as references evolve.
Important Notes
-
Internal Use Only:
Sdd Pipeline References is not intended for direct user invocation. It is designed for programmatic consumption by agents within the pipeline. -
Language and Format Consistency:
All schemas and rules are defined in Markdown or code-friendly formats, facilitating easy parsing and automation. -
Extensibility:
The references are modular and can be extended as the pipeline evolves. Custom agents or workflows can add their own schema documents, provided they follow the same conventions. -
Collaborative Maintenance:
Changes to reference documents should be peer-reviewed to prevent accidental breaking changes and to maintain pipeline stability.
By adhering to the standards set forth in Sdd Pipeline References, development teams can build more reliable, maintainable, and scalable agent-based pipelines.
More Skills You Might Like
Explore similar skills to enhance your workflow
Create Architecture
argument-hint: "[focus-area: full | layers | data-flow | api-boundaries | adr-audit] [--review full|lean|solo]"
Functions Skill
Execute serverless functions for data processing and API integration tasks
Modern Python
Automate and integrate modern Python workflows with up-to-date best practices
Plan CEO Review
Reviews plans from a CEO perspective by challenging premises and expanding product scope
Conducting Memory Forensics with Volatility
Performs memory forensics analysis using Volatility 3 to extract evidence of malware execution, process injection,
Epic Breakdown Advisor
Break down epics into user stories with Humanizing Work split patterns. Use when a backlog item is too large to estimate, sequence, or deliver safely