Implementation Templates¶
Purpose¶
These templates support both ES-106 — Implementation Readiness and ES-107 — AI-Assisted Implementation.
Use this family to move from accepted design to controlled implementation. The package separates readiness work from execution work so that teams do not jump into coding before repository state, work items, guardrails, AI use, verification, and review expectations are clear.
Recommended workflow¶
Design Readiness Summary
↓
ES-106 Implementation Readiness
↓
Repository + Branch + Work Item + Guardrail + AI + Verification Readiness
↓
Implementation Readiness Summary
↓
ES-107 AI-Assisted Implementation
↓
Implementation Log + AI Assistance Log + Evidence + Decisions + Risk Updates
↓
Work Item Completion Records
↓
Implementation Summary
↓
ES-108 Code Review and Integration
Copy location¶
Do not edit the reusable templates directly. Copy completed project versions into:
docs/project-workspace/implementation/
ES-106 readiness templates¶
Implementation Readiness Overview
Connect implementation readiness to accepted design and selected implementation scope.
02Repository Readiness Check
Confirm repository, build, test, evidence, branch, and review readiness.
03Branch and Workflow Plan
Define branch strategy, commit rules, pull requests, merge criteria, and review requirements.
04Implementation Work Items
Define traceable implementation work items, scope, done criteria, evidence, and verification.
05Implementation Guardrails
Define architecture, design, evidence, access, AI, security, and scope guardrails.
06AI-Use Plan
Define allowed, limited, prohibited, reviewed, and evidenced AI assistance.
07Verification Precheck
Identify existing checks, required new checks, evidence checks, access checks, and AI checks.
08Implementation Review Plan
Define what implementation review must cover and what evidence pull requests must provide.
09Implementation Readiness Summary
Summarize readiness to begin ES-107 implementation work.
ES-107 execution templates¶
ES-107 Guide
Use the execution templates for responsible AI-assisted implementation.
10Implementation Log
Record implementation activity, branches, PRs, design sources, checks, and status.
11AI Assistance Log
Record AI use, outputs used, verification performed, and rejected AI outputs.
12Implementation Evidence
Record code artifacts, evidence produced, guardrails checked, and verification performed.
13Implementation Decision Records
Record implementation-level decisions, rationale, consequences, and verification impact.
14Implementation Risk Updates
Record new, changed, retired, and carried-forward risks during implementation.
15Verification Notes
Record checks, results, failures, fixes, and known verification gaps.
16Work Item Completion Records
Record completion evidence, verification, reviewers, blocked items, and deferred items.
17Implementation Summary
Summarize implemented work, AI assistance, verification, gaps, and readiness for ES-108.
Completion expectations¶
| Question | Evidence |
|---|---|
| Is the team ready to implement? | implementation_readiness_summary.md |
| Is repository and workflow ready? | repository_readiness_check.md, branch_and_workflow_plan.md |
| What will be implemented? | implementation_work_items.md |
| What must not be broken? | implementation_guardrails.md |
| How may AI assist? | ai_use_plan.md |
| What checks are needed? | verification_precheck.md |
| How will implementation be reviewed? | implementation_review_plan.md |
| What happened during implementation? | implementation_log.md, ai_assistance_log.md |
| What evidence was produced? | implementation_evidence.md, verification_notes.md |
| What decisions and risks changed? | implementation_decision_records.md, implementation_risk_updates.md |
| Is the work ready for review/integration? | work_item_completion_records.md, implementation_summary.md |
Do not treat AI-generated implementation as completed work. AI output is only a candidate contribution until engineers review, test, integrate, and evidence it.
Implementation is not just writing code. It is preserving design intent, protecting guardrails, recording decisions, verifying behavior, and leaving enough evidence for review and future stewardship.