implementation-review
Automatically trigger review agents after task completion. Use when strategic-planner finishes planning tasks (calls plan-consultant) or when main agent completes coding tasks in /implement workflow (calls code-reviewer). Triggers on phrases like "plan complete", "implementation done", "coding finished", "ready for review".
When & Why to Use This Skill
The Implementation Review skill functions as an automated quality gate within the software development lifecycle. It intelligently monitors workflow progress and triggers specialized review agents—such as plan consultants and code reviewers—immediately upon the completion of planning or coding tasks. By orchestrating these handoffs, it ensures that every phase of the EPIC workflow meets rigorous standards for feasibility, security, and maintainability before proceeding.
Use Cases
- Automated Strategic Validation: Automatically invokes a plan-consultant agent once a strategic-planner finishes a project roadmap to identify risks and ensure architectural feasibility.
- Seamless Code Quality Gates: Triggers a code-reviewer agent immediately after the main agent completes implementation tasks, ensuring all new code adheres to project standards before the commit phase.
- Multi-Agent Workflow Synchronization: Manages the transition between different specialized agents (planners, developers, and reviewers) based on natural language triggers like 'plan complete' or 'coding finished', reducing manual intervention.
| name | implementation-review |
|---|---|
| description | Automatically trigger review agents after task completion. Use when strategic-planner finishes planning tasks (calls plan-consultant) or when main agent completes coding tasks in /implement workflow (calls code-reviewer). Triggers on phrases like "plan complete", "implementation done", "coding finished", "ready for review". |
Implementation Review Skill
Purpose
This skill ensures quality gates are enforced during the EPIC workflow by automatically invoking the appropriate review agents after task completion.
Review Triggers
1. Plan Review (Strategic-Planner Completion)
Trigger Conditions:
strategic-planneragent has completed its planning tasks- A strategy or implementation plan has been generated
- Keywords: "plan complete", "strategy ready", "planning done", "roadmap finalized"
Action: Delegate to plan-consultant agent for plan review
Review Focus:
- Plan completeness and feasibility
- Risk identification
- Alternative approaches consideration
- Alignment with project goals
2. Code Review (Main Agent Implementation Completion)
Trigger Conditions:
- Main agent has completed coding tasks in
/implementworkflow - Implementation code has been written (after TDD Green phase)
- Keywords: "implementation complete", "coding done", "feature implemented", "code ready"
Action: Delegate to code-reviewer agent for code review
Review Focus:
- Code quality and maintainability
- Security vulnerabilities
- Performance considerations
- Adherence to project standards
Workflow Integration
EPIC Workflow with Reviews:
[Explore] -> [Plan] -> PLAN REVIEW -> [Implement] -> CODE REVIEW -> [Commit]
| |
plan-consultant code-reviewer
Instructions
When Strategic-Planner Completes:
- Detect completion signal from strategic-planner agent
- Collect the generated plan/strategy document
- Invoke
plan-consultantagent with the plan for review - Report review findings back to main workflow
- If critical issues found, flag for plan revision before proceeding
When Main Agent Completes Coding:
- Detect completion of implementation tasks (TDD Green phase complete)
- Identify all files modified during implementation
- Invoke
code-revieweragent with the changed files - Report review findings back to main workflow
- If critical issues found, flag for code revision before commit
Agent Delegation
| Completion Event | Review Agent | Purpose |
|---|---|---|
strategic-planner done |
plan-consultant |
Validate implementation strategy |
| Main agent coding done | code-reviewer |
Validate code quality |
Review Report Format
After each review, expect a structured report:
## Review Summary
- **Reviewer:** [agent-name]
- **Rating:** [1-10]
- **Status:** [APPROVED / NEEDS REVISION / BLOCKED]
## Findings
- [Finding 1]
- [Finding 2]
## Recommendations
- [Recommendation 1]
- [Recommendation 2]
## Action Items
- [ ] [Required action if any]
Constraints
- DO NOT skip reviews - they are mandatory quality gates
- DO NOT proceed to next EPIC phase if review status is BLOCKED
- DO NOT modify code during review - only report findings
- Reviews should complete within reasonable time to not block workflow
Examples
Example 1: Plan Review Trigger
User: The strategic-planner has completed the implementation strategy for T001.