diff --git a/collections/gem-team.md b/collections/gem-team.md index 8d7b97be..c1c6c3db 100644 --- a/collections/gem-team.md +++ b/collections/gem-team.md @@ -26,14 +26,12 @@ recommended The Orchestrator is the coordination hub that coordinates multi-agent workflows, delegates tasks via runSubagent, and synthesizes results. It does not execute tasks directly but manages the overall workflow. This agent is ideal for: - - Coordinating complex multi-agent workflows - Managing task delegation and parallel execution - Synthesizing results from multiple agents - Maintaining plan.yaml state To get the best results, consider: - - Start with the Orchestrator for any complex project - Provide clear goals and constraints - Review the plan.yaml before execution @@ -48,14 +46,12 @@ recommended The Researcher gathers codebase context, identifies relevant files/patterns, and returns structured findings. It is typically invoked by the Orchestrator with a specific focus area. This agent is ideal for: - - Understanding codebase structure and patterns - Identifying relevant files for a specific feature - Gathering context before making changes - Researching technical dependencies To get the best results, consider: - - Specify a clear focus area or question - Provide context about what you're trying to achieve - Use multiple Researchers in parallel for different areas @@ -69,14 +65,12 @@ recommended The Planner creates DAG-based plans with pre-mortem analysis, presents for approval, and iterates on feedback. It synthesizes research findings into a structured plan. This agent is ideal for: - - Breaking down complex goals into atomic tasks - Creating task dependencies (DAG) - Running pre-mortem analysis to identify risks - Getting approval before execution To get the best results, consider: - - Provide clear research findings from the Researcher - Review the plan carefully before approving - Ask for iterations if the plan is not optimal @@ -91,14 +85,12 @@ recommended The Implementer executes TDD code changes, ensures verification, and maintains quality. It follows strict TDD discipline with verification commands. This agent is ideal for: - - Implementing features with TDD discipline - Writing tests first, then code - Ensuring verification commands pass - Maintaining code quality To get the best results, consider: - - Always provide verification commands - Follow TDD: red, green, refactor - Check get_errors after every edit @@ -113,14 +105,12 @@ optional The Chrome Tester automates browser testing and UI/UX validation via Chrome DevTools. It requires Chrome DevTools MCP server. This agent is ideal for: - - Automated browser testing - UI/UX validation - Capturing screenshots and snapshots - Testing web applications To get the best results, consider: - - Have Chrome DevTools MCP server installed - Provide clear test scenarios - Use snapshots for debugging @@ -135,14 +125,12 @@ optional The DevOps agent manages containers, CI/CD pipelines, and infrastructure deployment. It handles infrastructure as code and deployment automation. This agent is ideal for: - - Setting up CI/CD pipelines - Managing containers (Docker, Kubernetes) - Infrastructure deployment - DevOps automation To get the best results, consider: - - Provide clear infrastructure requirements - Use IaC best practices - Test pipelines locally @@ -157,14 +145,12 @@ recommended The Reviewer is a security gatekeeper for critical tasks. It applies OWASP scanning, secrets detection, and compliance verification. This agent is ideal for: - - Security code reviews - OWASP Top 10 scanning - Secrets and PII detection - Compliance verification To get the best results, consider: - - Use for all critical security changes - Review findings carefully - Address all security issues @@ -179,14 +165,12 @@ optional The Documentation Writer generates technical docs, diagrams, and maintains code-documentation parity. This agent is ideal for: - - Generating technical documentation - Creating diagrams - Keeping docs in sync with code - API documentation To get the best results, consider: - - Provide clear context and requirements - Review generated docs for accuracy - Update docs with code changes @@ -194,4 +178,4 @@ To get the best results, consider: --- -*This collection includes 8 curated items for **Gem Team Multi-Agent Orchestration**.* +*This collection includes 8 curated items for **Gem Team Multi-Agent Orchestration**.* \ No newline at end of file