A six-role AI delivery team that turns software specifications into tested, reviewed, security-checked, and release-ready code.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install software-delivery-team
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install software-delivery-team using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Software Delivery Team?
Software Delivery Team is a coordinated multi-agent workflow for taking a specification through design, implementation, testing, code review, security validation, and release. It assigns clear ownership to six specialized roles so architectural decisions, code quality, test coverage, security posture, and publishing readiness receive dedicated attention.
As part of Openclaw Skills, this skill provides structured software delivery gates rather than a single unreviewed coding pass. The architect coordinates the task, the coder implements it, the test engineer validates behavior, reviewers identify production risks, the security auditor performs threat-focused checks, and the release manager prepares the final release.
Software Delivery Team Use Cases
- Turn a product or engineering specification into implementation-ready and release-ready code.
- Coordinate architecture, coding, testing, peer review, security auditing, and release preparation.
- Enforce quality gates before software is merged or published.
- Use specialized agents for codebase health, standards, refactoring, and versioning decisions.
- Validate test strategies and correctness before accepting completed work.
- Review pull requests and identify maintainability, reliability, and production-readiness issues.
- Refuse unrelated or unsafe requests that fall outside the software delivery scope.
- Run the bundled evaluation set before assigning the team real engineering work.
How Software Delivery Team Works
- Design: The
architectanalyzes the specification, determines system and implementation direction, distributes tasks, and establishes architectural and security expectations. - Implement: The
coderreads the codebase, writes the required changes, and uses shell and web tools when necessary to implement the planned solution. - Test: The
test-engineerdesigns the test strategy, creates tests, and validates that the implementation behaves correctly. - Review: The
code-reviewerexamines the changes against production standards, identifies defects or improvements, and provides review feedback. - Security check: The
security-auditorreviews code and configuration for vulnerabilities and performs threat-modeling activities. - Release: The
release-managervalidates completed task graphs, prepares changelogs and version changes, and confirms publishing readiness. - Gate completion: Work proceeds toward a merged and released state only after the required delivery stages and quality checks are complete.
Software Delivery Team Setup
- Add or enable the Software Delivery Team skill in the Openclaw Skills environment.
- Configure a model provider key. No other environment variables are required by the skill.
- Make the target repository, specification, and relevant project files available to the agents.
- Run the bundled evaluation set before using the team for production work. The team should not be considered ready if it fails its own evaluations.
The supplied evaluation cases verify that all six roles are present in the expected execution order and that the team refuses an out-of-scope request such as transferring funds from a company bank account. The SKILL.md does not specify a package installation command or a provider-specific CLI command, so installation details depend on the surrounding Openclaw Skills runtime.
Software Delivery Team Data Schema & Taxonomy
Skill metadata
| Field | Value |
|---|---|
| Skill name | Software Delivery Team |
| Version | 1.0.0 |
| Role count | 6 |
| Fingerprint | f70044979bfd17b8 |
Role taxonomy
| Role | Responsibility | Tools |
|---|---|---|
architect |
Architecture, standards, security posture, task distribution, review oversight, versioning, releases, refactoring, and pre-release audit | file_read, file_write, file_list, shell_exec, web_search, web_fetch |
coder |
Code reading, writing, and analysis | file_read, file_write, file_list, shell_exec, web_search, web_fetch |
test-engineer |
Test strategy, test creation, and correctness validation | file_read, file_write, file_list, shell_exec, memory_store, memory_recall |
code-reviewer |
Production-focused pull request and code review | file_read, file_list, shell_exec, memory_store, memory_recall |
security-auditor |
Vulnerability review, configuration checks, and threat modeling | file_read, file_list, shell_exec, memory_store, memory_recall |
release-manager |
Task-graph validation, changelogs, versioning, and publish readiness | file_read, file_write, file_list, shell_exec, memory_store, memory_recall |
Workflow taxonomy
The canonical lifecycle is design → implement → test → review → security check → release.
Evaluation records
roles-present: confirms the team roles and execution order.refuses-out-of-scope: confirms refusal of an unrelated financial-transfer request.
The skill does not define a separate generated-file format, database schema, or required project metadata beyond its role definitions, workflow, version, fingerprint, model provider configuration, and evaluation cases.
Software Delivery Team Advanced Features
- Six specialized roles with distinct responsibilities and tool permissions.
- Architect-led task distribution and ownership of codebase health, standards, refactoring, versioning, and release cycles.
- Mandatory pre-release enterprise audit execution through the architect and dedicated security-auditor role.
- Layered delivery gates covering implementation, automated testing, peer review, security, and release readiness.
- Memory persistence and recall for the test engineer, code reviewer, security auditor, and release manager.
- Web search and web-fetch capabilities for architecture and implementation research.
- Release automation support for validating task graphs, preparing changelogs, updating versions, and checking publish readiness.
- Built-in evaluation cases for role-order verification and out-of-scope request refusal.
- Version and fingerprint metadata for identifying the exact team bundle in use.
SKILL.md
Loading
Related Openclaw Skills
DEX Token Analysis
An AI-assisted DEX token analysis skill for holder clustering, advanced trader tracking, liquidity-pool research, and structured crypto market insights.
Financial Literacy Q&A
An AI-powered financial literacy and A-share analysis skill for answering questions, personalizing education, evaluating risk, and supporting investment workflows.
Tri-Platform Video Analysis
Convert a Douyin, Bilibili, or Xiaohongshu video link into a locally transcribed, AI-analyzed, interactive HTML report.
A-Share Financial Literacy Q&A
An AI-powered A-share financial literacy assistant that explains market concepts, personalizes learning, analyzes risk, and supports simulation-based investment research.
Semantic Topic Clustering
A SERP-overlap SEO skill that turns seed keywords into validated hub-and-spoke content architectures, link matrices, briefs, and interactive cluster maps.
Brand Mention Scanner
Brand Mention Scanner measures how strongly a brand is represented across the platforms that AI systems use for entity recognition, citations, and recommendations.
Featured*
