Implement E24-T2 release assistant engine backend (#111)

Author: dmoliveira

CHANGELOG.md

@@ -153,6 +153,7 @@ All notable changes to this project are documented in this file.
 - Integrated `pr-review` checks into unified `/doctor`, updated installer self-check/hints for PR review workflows, and marked Task 23.3 complete in the roadmap.
 - Expanded pr-review verification coverage for risk-detection false-positive control and missing-evidence behavior, and marked Epic 23 Task 23.4/exit criteria complete in the roadmap.
 - Added release-train policy contract covering deterministic preflight gates, semantic-version blocking rules, and partial-failure rollback reason codes; marked Epic 24 Task 24.1 complete in the roadmap.
+- Added `release_train_engine.py` backend with deterministic release preflight diagnostics, draft-note generation from git history, and confirmation-gated publish dry-run flow; marked Epic 24 Task 24.2 complete in the roadmap.
 
 ## v0.2.0 - 2026-02-12
 

IMPLEMENTATION_ROADMAP.md

@@ -60,7 +60,7 @@ Use this map to avoid overlapping implementations.
 | E21 | Bounded Loop Mode Presets | merged | Medium | E22, E28 | TBD | Merged into E22/E28 loop controls |
 | E22 | Autoflow Unified Orchestration Command | done | High | E14, E15, E17, E19, E20 | TBD | One command for plan-run-resume-report lifecycle |
 | E23 | PR Review Copilot | done | High | E3 | bd-u6t | Pre-PR quality, output, and risk review automation |
-| E24 | Release Train Assistant | in_progress | High | E14, E23 | bd-219 | Validate, draft, and gate releases reliably |
+| E24 | Release Train Assistant | in_progress | High | E14, E23 | bd-1u4 | Validate, draft, and gate releases reliably |
 | E25 | Incident Hotfix Mode | planned | Medium | E20, E22 | TBD | Constrained emergency workflow with strict safety |
 | E26 | Repo Health Score and Drift Monitor | planned | Medium | E9, E12, E20 | TBD | Operational visibility and continuous diagnostics |
 | E27 | Knowledge Capture from Completed Tasks | planned | Medium | E9, E14, E23 | TBD | Convert delivered work into reusable team memory |
@@ -851,10 +851,11 @@ Every command-oriented epic must ship all of the following:
   - [x] Subtask 24.1.2: Define semantic version rules and validation
   - [x] Subtask 24.1.3: Define rollback strategy for partial release failures
   - [x] Notes: Added `instructions/release_train_policy_contract.md` with deterministic preflight gates, semantic version mapping rules, command reason-code requirements, and partial-failure rollback strategy for `/release-train` workflows.
-- [ ] Task 24.2: Implement release assistant engine
-  - [ ] Subtask 24.2.1: Add preflight checks and blocking diagnostics
-  - [ ] Subtask 24.2.2: Generate draft release notes from merged changes
-  - [ ] Subtask 24.2.3: Add dry-run publish flow with explicit confirmation step
+- [x] Task 24.2: Implement release assistant engine
+  - [x] Subtask 24.2.1: Add preflight checks and blocking diagnostics
+  - [x] Subtask 24.2.2: Generate draft release notes from merged changes
+  - [x] Subtask 24.2.3: Add dry-run publish flow with explicit confirmation step
+  - [x] Notes: Added `scripts/release_train_engine.py` with `status|prepare|draft|publish|doctor` flows, deterministic preflight reason-code diagnostics, changelog/version gating, release-note draft generation from git history, and confirmation-gated publish simulation.
 - [ ] Task 24.3: Command integration
   - [ ] Subtask 24.3.1: Add `/release-train status|prepare|draft|publish`
   - [ ] Subtask 24.3.2: Integrate with existing `make release-check` and changelog flow

README.md

@@ -267,6 +267,13 @@ Task 24.1 release policy contract notes:
 - semantic-version gating now defines deterministic patch/minor/major validation and breaking-change mismatch blocking.
 - rollback policy now defines partial-failure handling with explicit reason codes and follow-up guidance.
 
+Task 24.2 release assistant engine notes:
+
+- backend module: `scripts/release_train_engine.py` with `status`, `prepare`, `draft`, `publish`, and `doctor` command flows.
+- `prepare` emits deterministic `reason_codes` and remediation for clean-tree, branch, validation, changelog, and semver gating checks.
+- `draft` composes release-note entries from git history since the latest (or provided) tag.
+- `publish` enforces readiness and explicit confirmation, with dry-run and rollback action metadata.
+
 ## Installed plugin stack 🔌
 
 - `@mohak34/opencode-notifier@latest` - desktop and sound alerts for completion, errors, and permission prompts.
@@ -1253,6 +1260,7 @@ For your LangGraph setup, default endpoint target is `http://localhost:3000/open
 - `scripts/autoflow_adapter.py` - orchestration adapter for `/autoflow` transition and explain planning
 - `scripts/autoflow_command.py` - unified `/autoflow` command surface with dry-run and kill-switch controls
 - `scripts/budget_command.py` - backend script for `/budget`
+- `scripts/release_train_engine.py` - release-train backend engine for preflight, draft, and publish gating
 - `scripts/todo_command.py` - backend script for `/todo`
 - `scripts/resume_command.py` - backend script for `/resume`
 - `scripts/safe_edit_adapters.py` - semantic safe-edit adapter and validation helpers