@jakkrichm/create-nexus-devflow 2.0.16 → 2.0.18

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (67) hide show
  1. package/dist/bin/create-nexus-devflow.d.ts +2 -1
  2. package/dist/bin/create-nexus-devflow.js +81 -3
  3. package/dist/bin/create-nexus-devflow.js.map +1 -1
  4. package/dist/lib/current-work.d.ts +2 -3
  5. package/dist/lib/current-work.js +93 -66
  6. package/dist/lib/current-work.js.map +1 -1
  7. package/dist/lib/uninstall.d.ts +20 -0
  8. package/dist/lib/uninstall.js +129 -0
  9. package/dist/lib/uninstall.js.map +1 -0
  10. package/dist/scripts/prepare-template.js +120 -10
  11. package/dist/scripts/prepare-template.js.map +1 -1
  12. package/package.json +1 -1
  13. package/template/.agents/skills/00-discover/SKILL.md +2 -2
  14. package/template/.agents/skills/10-define/SKILL.md +24 -140
  15. package/template/.agents/skills/20-spec/SKILL.md +2 -2
  16. package/template/.agents/skills/30-plan/SKILL.md +2 -2
  17. package/template/.agents/skills/50-verify/SKILL.md +2 -2
  18. package/template/.agents/skills/60-report/SKILL.md +1 -1
  19. package/template/.agents/skills/70-release/SKILL.md +19 -143
  20. package/template/.agents/skills/check/SKILL.md +10 -10
  21. package/template/.agents/skills/ci/SKILL.md +1 -1
  22. package/template/.agents/skills/complete/SKILL.md +49 -28
  23. package/template/.agents/skills/deploy/SKILL.md +1 -1
  24. package/template/.agents/skills/devflow/SKILL.md +5 -5
  25. package/template/.agents/skills/feature/SKILL.md +25 -20
  26. package/template/.agents/skills/fix/SKILL.md +25 -19
  27. package/template/.agents/skills/implement/SKILL.md +10 -10
  28. package/template/.agents/skills/insight/SKILL.md +1 -1
  29. package/template/.agents/skills/overview/SKILL.md +3 -3
  30. package/template/.agents/skills/report-html/SKILL.md +3 -3
  31. package/template/.agents/skills/security-review/SKILL.md +1 -1
  32. package/template/.agents/skills/spec/SKILL.md +7 -6
  33. package/template/.agents/skills/spec-driven-development/SKILL.md +2 -2
  34. package/template/.agents/skills/test/SKILL.md +1 -1
  35. package/template/.claude/skills/00-discover/SKILL.md +2 -2
  36. package/template/.claude/skills/10-define/SKILL.md +24 -140
  37. package/template/.claude/skills/20-spec/SKILL.md +2 -2
  38. package/template/.claude/skills/30-plan/SKILL.md +2 -2
  39. package/template/.claude/skills/50-verify/SKILL.md +2 -2
  40. package/template/.claude/skills/60-report/SKILL.md +1 -1
  41. package/template/.claude/skills/70-release/SKILL.md +19 -143
  42. package/template/.claude/skills/check/SKILL.md +10 -10
  43. package/template/.claude/skills/ci/SKILL.md +1 -1
  44. package/template/.claude/skills/complete/SKILL.md +49 -28
  45. package/template/.claude/skills/deploy/SKILL.md +1 -1
  46. package/template/.claude/skills/devflow/SKILL.md +5 -5
  47. package/template/.claude/skills/feature/SKILL.md +25 -20
  48. package/template/.claude/skills/fix/SKILL.md +25 -19
  49. package/template/.claude/skills/implement/SKILL.md +10 -10
  50. package/template/.claude/skills/insight/SKILL.md +1 -1
  51. package/template/.claude/skills/overview/SKILL.md +3 -3
  52. package/template/.claude/skills/report-html/SKILL.md +3 -3
  53. package/template/.claude/skills/security-review/SKILL.md +1 -1
  54. package/template/.claude/skills/spec/SKILL.md +7 -6
  55. package/template/.claude/skills/spec-driven-development/SKILL.md +2 -2
  56. package/template/.claude/skills/test/SKILL.md +1 -1
  57. package/template/AGENTS.md +15 -59
  58. package/template/devflow/context/ai-interaction.md +110 -23
  59. package/template/devflow/context/coding-standards.md +123 -32
  60. package/template/devflow/context/current-feature.md +3 -0
  61. package/template/devflow/context/current-stage.md +6 -6
  62. package/template/devflow/context/findings.md +10 -10
  63. package/template/devflow/context/project-overview.md +18 -32
  64. package/template/devflow/history/HISTORY.md +10 -25
  65. package/template/devflow/ideas.md +9 -5
  66. package/template/devflow/reference/running-id-contract.md +39 -39
  67. package/template/devflow/runs/.gitkeep +0 -0
@@ -1,35 +1,122 @@
1
- # AI Interaction Guidelines for DevFlow
1
+ # AI Interaction Guidelines for Nexus-DevFlow
2
2
 
3
- > **DevFlow is an agentic workflow layer**, overlaying on top of scaffolded or existing codebases.
3
+ > **Nexus-DevFlow is an agentic workflow layer**, overlaying on top of scaffolded or existing codebases. Never run a framework scaffolder inside an initialized DevFlow directory.
4
4
 
5
- ## Communication & Interaction
5
+ ---
6
6
 
7
- - Be concise and direct in communication.
8
- - Explain technical trade-offs and non-obvious design decisions briefly.
9
- - Always confirm before performing major refactors or destructive operations.
10
- - Maintain markdown-first evidence for all stage activities.
7
+ ## 1. Communication & Principles
11
8
 
12
- ## Artifact Language
9
+ - **Be Concise and Direct**: State conclusions, status, and findings first; provide supporting context afterward.
10
+ - **Explain Non-Obvious Decisions Briefly**: Highlight architectural trade-offs, edge-case rationale, or safety considerations in 1–2 sentences.
11
+ - **Ask Before Destructive or Architectural Changes**: Always obtain explicit confirmation before deleting files, executing major refactors, or altering public interfaces.
12
+ - **Don't Add Unplanned Scope**: Stick strictly to the Acceptance Criteria defined in the spec. Avoid adding "nice-to-have" features that were not requested.
13
+ - **Preserve Existing Codebase Patterns**: Respect existing file structure, typing patterns, naming conventions, and deep module boundaries.
13
14
 
14
- - Default language for user-facing artifacts and summaries is Thai (`th`), while code, paths, and identifiers remain in English.
15
- - Use exact schema headings when validators check for specific contract sections.
15
+ ---
16
16
 
17
- ## Output Formatting
17
+ ## 2. Output Formatting
18
18
 
19
- - Use GitHub-style markdown for clean scanning.
20
- - Use lists for steps, options, and findings.
21
- - Use tables for comparative matrices or stage summaries.
22
- - Use backticks for paths, filenames, identifiers, and CLI commands.
19
+ Format every response for fast scanning and readability:
23
20
 
24
- ## DevFlow Timeline Lifecycle
21
+ - **Real Markdown, Not Prose Walls**: Use bold labels, concise bullet points, and blank lines between blocks.
22
+ - **Enumerations Are Lists**: Numbered or bulleted lists for sequential steps or findings, never inline runs crammed into paragraphs.
23
+ - **Tables for Comparative Matrices**: Use markdown tables when comparing status, options, test results, or trade-offs.
24
+ - **Backticks for Code References**: Wrap file paths, variable names, functions, CLI flags, and commands in backticks (e.g. `current-feature.md`, `npm run check`).
25
+ - **Clickable File Links**: Use GitHub-style markdown links with `file://` scheme (e.g. `[current-feature.md](file:///d:/path/to/current-feature.md)`).
26
+ - **Lead With the Result**: State pass/fail status or completed action before presenting logs.
27
+
28
+ ---
29
+
30
+ ## 3. Dual-Track Workflow Lifecycle (The 3-Pillars Model)
25
31
 
26
32
  ```text
27
- /00-Discover -> /10-Define -> /20-Spec -> /30-Plan -> /40-Execute -> /50-Verify -> /60-Report -> /70-Release
33
+ devflow/
34
+ ├── 🔮 ideas.md # [Future] Idea Inbox
35
+ ├── ⚡ context/ # [Present] Living Spec (current-feature.md) & Active State
36
+ └── 📦 history/ # [Past] features/, fixes/, rollbacks/, and HISTORY.md
28
37
  ```
29
38
 
30
- - Run `/00-Discover` to explore requests under a Discovery ID.
31
- - Run `/10-Define` to set delivery boundaries and allocate Running IDs.
32
- - Run `/20-Spec` and `/30-Plan` to establish formal delivery specifications and execution plans.
33
- - Run `/40-Execute` for step-by-step code implementation with evidence.
34
- - Run `/50-Verify`, `/60-Report`, and `/70-Release` for quality verification, reporting, and release packaging.
35
- - Use **Fast-Track / Quick-Fix mode** for minor hotfixes to skip heavy ceremony while keeping test evidence.
39
+ ### 🏎️ Track 1: Fast-Track (Blueprint Mode 4 Steps)
40
+ *Recommended for 85% of daily engineering work (features, bug fixes, UI improvements, iterative refactoring).*
41
+
42
+ The entire lifecycle is driven by the **Single Living Spec (`devflow/context/current-feature.md`)**:
43
+
44
+ 1. **Spec (`/feature` or `/fix`)**:
45
+ - Checks **Single Active Run Guardrail** (rejects if an uncompleted task is still active).
46
+ - Analyzes request (or consumes `IDEA-xxx` from `devflow/ideas.md`).
47
+ - Allocates sequential ID without prefix (e.g. `022-{slug}`) and creates branch `feature/{xxx-slug}` or `fix/{xxx-slug}`.
48
+ - Generates `devflow/context/current-feature.md` containing **Section 1 (Scope & AC)**, **Section 2 (Plan & Test Strategy)**, and **Section 3 (Checklist)**.
49
+ 2. **Implement (`/implement`)**:
50
+ - Executes checklist tasks incrementally one small diff at a time using **TDD (Red-Green-Refactor)**.
51
+ - Updates `## 4. Implementation Record` and marks tasks `- [x]` in `current-feature.md`.
52
+ 3. **Check (`/check`)**:
53
+ - Senior QA multi-lane verification (Lane 1: Typecheck/Lint, Lane 2: Test Suites, Lane 3: Manual Proof).
54
+ - Records empirical proof under `## 5. Verification Evidence` in `current-feature.md`.
55
+ 4. **Complete (`/complete`)**:
56
+ - Final safety pass, updates `## 6. Release & Handoff` digest in `current-feature.md`.
57
+ - Automatically archives `current-feature.md` ➔ `devflow/history/{features|fixes|rollbacks}/{xxx-slug}.md`.
58
+ - Appends resolved findings and cleans `findings.md`.
59
+ - Resets `current-feature.md` back to the idle stub.
60
+ - Performs Git squash-merge into `main`, updates `devflow/history/HISTORY.md`, and sets workspace to Idle.
61
+
62
+ ---
63
+
64
+ ### 🏗️ Track 2: Deep-Track (Architect Mode — 8 Steps)
65
+ *Recommended for large architectural epics, database migrations, security audits, and multi-agent coordination.*
66
+
67
+ ```text
68
+ 00-discover ➔ 10-define ➔ 20-spec ➔ 30-plan ➔ 40-execute ➔ 50-verify ➔ 60-report ➔ 70-release
69
+ ```
70
+
71
+ 1. `00-discover`: Explore request before delivery commitment (`DISC-YYYYMMDD-NNN`).
72
+ 2. `10-define`: Turn approved discovery into bounded delivery run in `devflow/context/current-run/10-define.md`.
73
+ 3. `20-spec`: Formalize markdown delivery contract & acceptance criteria (`20-spec.md`).
74
+ 4. `30-plan`: Breakdown spec into executable tasks with test decisions (`30-plan.md` + checklists).
75
+ 5. `40-execute`: Incremental task execution behind review gates (`40-execute.md`).
76
+ 6. `50-verify`: Senior QA review & multi-lane verification checks (`50-verify.md`).
77
+ 7. `60-report`: Standardized markdown delivery digest (`60-report.md`).
78
+ 8. `70-release`: Release packaging, git merge, archives `devflow/context/current-run/` ➔ `devflow/history/{features|fixes|rollbacks}/{xxx-slug}/`, and closes the run.
79
+
80
+ ---
81
+
82
+ ## 4. Standalone HTML Reporting Policy
83
+
84
+ > [!IMPORTANT]
85
+ > **No Auto-Generated HTML**: Mainline stages (`/complete` and `60-report`) strictly output Markdown only.
86
+ > When an interactive web dashboard is desired for presentation or sharing, invoke the standalone companion command:
87
+ > `/report:html` (or `npm run report:html -- {ID}`).
88
+
89
+ ---
90
+
91
+ ## 5. Resuming After Context Clear
92
+
93
+ Progress lives in persistent files, not in transient chat history:
94
+
95
+ - In Fast-Track: `devflow/context/current-feature.md` maintains ticked checklist boxes `- [x]`.
96
+ - In Deep-Track: `devflow/context/current-run/` maintains stage markdown files.
97
+ - In Git: Commits, branches, and working tree maintain the code history.
98
+ - When starting a fresh session after a context clear, run `devflow` or inspect `current-stage.md` to pick up immediately from the next pending step.
99
+
100
+ ---
101
+
102
+ ## 6. Single Active Run Guardrail (One Thing at a Time)
103
+
104
+ - Only one active run is allowed at a time across both Fast-Track and Deep-Track.
105
+ - The AI will actively block opening a new feature or fix until the current one is completed with `/complete` or `70-release` (or explicitly rolled back/cancelled).
106
+
107
+ ---
108
+
109
+ ## 7. Branching & Git Conventions
110
+
111
+ - **Branch Naming**: `feature/{xxx-slug}` or `fix/{xxx-slug}`.
112
+ - **Commit Messages**: Conventional imperative format (e.g. `feat(uninstall): add clean eject CLI command`, `fix(parser): handle undefined metadata field`).
113
+ - **No AI Attribution in Commits**: Never include "Generated with AI" or agent metadata in Git commit logs.
114
+ - **Explicit Approval for Push & Deploy**: Merge approval is strictly separate from consent to `git push` to remote repositories or deploy to production.
115
+
116
+ ---
117
+
118
+ ## 8. Autopilot Policy
119
+
120
+ - `autopilot` is an explicit opt-in command (`/autopilot`). Never suggest it as the default next action.
121
+ - When invoked, it runs one bounded spec/plan/implement/verify pass.
122
+ - Autopilot **MUST stop** before `/complete`, merge, push, deploy, or any destructive action.
@@ -1,40 +1,131 @@
1
- # Coding Standards & Guidelines
1
+ # Coding Standards & Engineering Guidelines
2
2
 
3
- General engineering standards for code, tests, and documentation in Nexus-DevFlow.
3
+ > Engineering standards, architecture conventions, testing policies, and code quality guidelines for **Nexus-DevFlow**.
4
4
 
5
- ## 1. Code Quality & Architecture
5
+ ---
6
6
 
7
- - Write clear, self-documenting code with single-responsibility functions.
8
- - Avoid unnecessary complexity (prefer simple solutions over indirect abstractions).
9
- - Maintain type safety and static validation across module boundaries.
10
- - Treat `project-overview.md` as the single source of truth for concrete data models and entity definitions.
7
+ ## 1. TypeScript & Runtime Standards
11
8
 
12
- ## 2. Testing & Empirical Proof Standards
9
+ - **Strict Mode Enabled**: Always adhere to strict TypeScript compilation (`"strict": true` in `tsconfig.json`).
10
+ - **No `any` Types**: Prohibit the use of `any`. Use precise interface definitions, generics, or `unknown` with runtime type narrowing.
11
+ - **Type Definitions**: Define clear interfaces and types for all function options, CLI arguments, parse results, and data payloads.
12
+ - **Type Inference vs. Explicit Types**: Use type inference for straightforward local variables, and explicit return types for public functions, exports, and complex utilities.
13
+ - **ESM-First Architecture**: Use native Node.js ESM (`"type": "module"`). Always use explicit `node:` protocol imports (e.g. `node:fs`, `node:path`, `node:os`, `node:child_process`).
13
14
 
14
- - **Unit Test Mandate**: Any logic modification or bug fix must be accompanied by new or updated unit tests.
15
- - **Empirical Proof Contract**: Never claim "passed", "verified", or "working" without citing exact command outputs, test run reports, route responses, logs, or screenshots.
16
- - **Manual Try Guide**: Every delivery run must provide a human-testable Try Guide ("Where to go", "What to click", "What to expect") for manual validation.
15
+ ---
17
16
 
18
- ## 3. Findings Ledger & Quality Gates (`findings.md`)
17
+ ## 2. CLI Architecture & Engineering Principles
19
18
 
20
- - All audit, security, and verification issues must be tracked in `devflow/context/findings.md`.
19
+ - **Separation of Concerns (Deep Modules)**:
20
+ - Keep CLI entry points (`bin/create-nexus-devflow.ts`) thin: handle argument parsing, option normalization, and terminal formatting.
21
+ - Encapsulate all core business logic, filesystem operations, and parsing inside modular libraries (`lib/current-work.ts`, `lib/findings.ts`, `lib/git.ts`, `lib/uninstall.ts`, `lib/update.ts`).
22
+ - **Safety Flags & Idempotency**:
23
+ - Destructive or mutating operations (e.g. `uninstall`, `update`, `install`) must support safety flags:
24
+ - `--dry-run`: Preview actions and affected files without modifying the disk.
25
+ - `-y`, `--yes`, `-f`, `--force`: Non-interactive mode for CI/CD automation.
26
+ - `--json`: Machine-readable structured output for integrations.
27
+ - **Cross-Platform Compatibility**:
28
+ - Never assume POSIX-only paths or Windows-only backslashes. Always use `path.join()`, `path.resolve()`, or normalize paths with forward slashes for URLs and identifiers.
29
+ - Handle Windows permission quirks (EPERM, EBUSY) with graceful retries or clear explanatory messages.
30
+ - **Terminal UX & ANSI Formatting**:
31
+ - Provide clear, high-contrast, and aesthetic terminal outputs.
32
+ - Automatically respect `process.stdout.isTTY` and support `--no-color` / `NO_COLOR` environment variables.
33
+
34
+ ---
35
+
36
+ ## 3. File Organization & Directory Structure
37
+
38
+ ```text
39
+ nexus-devflow/
40
+ ├── .agents/skills/ # Codex & Google Antigravity skill definitions
41
+ ├── .claude/skills/ # Claude Code mirrored skill adapters
42
+ ├── .nexus/ # Metadata tracking & upstream baseline ledger
43
+ ├── devflow/ # Framework workspace context, runs, and discoveries
44
+ │ ├── context/ # Living source-of-truth context files
45
+ │ ├── discoveries/ # Pre-delivery discovery records (00-discover.md)
46
+ │ ├── runs/ # Running delivery artifacts (current-feature.md / 00-70)
47
+ │ ├── history/ # Master delivery archive (HISTORY.md)
48
+ │ └── ideas.md # Idea Inbox and backlog
49
+ ├── packages/
50
+ │ └── create-nexus-devflow/ # Distribution npm package source
51
+ │ ├── bin/ # CLI executable binaries
52
+ │ ├── lib/ # Modular domain libraries
53
+ │ └── test/ # Automated unit tests (*.test.ts)
54
+ └── scripts/ # Maintainer verification & automation scripts
55
+ ```
56
+
57
+ ---
58
+
59
+ ## 4. Naming Conventions
60
+
61
+ - **Files & Directories**:
62
+ - Skills and command directories: `kebab-case` (e.g. `00-discover`, `report-html`, `sync-upstream`).
63
+ - Source modules and scripts: `kebab-case.ts` / `kebab-case.mjs` (e.g. `current-work.ts`, `check-devflow.ts`).
64
+ - Test files: `[module].test.ts` (e.g. `status.test.ts`, `uninstall.test.ts`).
65
+ - **Identifiers**:
66
+ - Functions & methods: `camelCase` (e.g. `parseCurrentWork`, `applyUninstall`, `readProjectStatus`).
67
+ - Variables & properties: `camelCase`.
68
+ - Types & Interfaces: `PascalCase` without prefixes like `I` or `T` (e.g. `ProjectStatus`, `RunWorkSummary`).
69
+ - Constants & Enums: `SCREAMING_SNAKE_CASE` (e.g. `LEGACY_FEATURE_PATH`, `DEFAULT_ADAPTERS`).
70
+
71
+ ---
72
+
73
+ ## 5. Testing & Empirical Proof Standards
74
+
75
+ Testing is a core quality gate in Nexus-DevFlow, not an afterthought:
76
+
77
+ - **Unit Test Mandate**: Any new feature, modified logic, parser improvement, or bug fix **MUST ship with automated unit tests** in the same diff.
78
+ - **Test Framework**: Use Node.js native test runner executed via `tsx --test test/*.test.ts` under `packages/create-nexus-devflow/`.
79
+ - **Test Design (AAA Pattern)**:
80
+ - Structure each test case cleanly: **Arrange** (setup fixtures/mock directories), **Act** (execute function), **Assert** (verify invariants).
81
+ - Use isolated temporary directories (`fs.mkdtemp` in `os.tmpdir()`) and ensure cleanup in `finally` blocks.
82
+ - **Empirical Proof Contract**:
83
+ - Never claim a task is "working", "tested", or "verified" without providing concrete empirical proof (exact command executed, terminal output, pass/fail counts, exit code).
84
+ - **3-Lane Verification Matrix**:
85
+ - **Lane 1 (Type & Syntax Safety)**: `tsc --noEmit` (0 type errors).
86
+ - **Lane 2 (Automated Test Suites & Evals)**: `npm test` (Unit tests 100% pass) + `npm run test:routing` (Skill routing accuracy).
87
+ - **Lane 3 (Manual / Scenario Proof)**: Concrete walkthrough steps ("Where to go", "What to run", "What to expect").
88
+
89
+ ---
90
+
91
+ ## 6. Findings Ledger & Quality Gates (`findings.md`)
92
+
93
+ - All quality defects, security findings, or regression issues identified during review must be logged in `devflow/context/findings.md`.
21
94
  - **Finding State Machine**:
22
- - `open`: Confirmed defect not yet repaired.
23
- - `fixed`: Repaired in code, pending verification.
24
- - `closed`: Verified in `50-verify` as completely resolved without regressions.
25
- - `accepted`: Waived with recorded user rationale.
26
- - `invalid`: Proven non-issue with concrete evidence.
27
- - **P0/P1 Blockers**: P0 and P1 findings in `open` or `fixed` status block `70-release` unconditionally. `fixed` status must be promoted to `closed` via `50-verify` re-examination.
28
-
29
- ## 4. Git Workflow & Release Safety
30
-
31
- - Commit messages follow conventional/imperative format (e.g., `feat(RUN-007): ...`).
32
- - Feature branches follow `feature/{slug}-{running-id}` or `fix/{slug}-{running-id}`.
33
- - **2-Stage Release Approvals**: Explicit consent to merge into `main` is strictly separate from consent to `git push` to remote repositories or deploy to production.
34
-
35
- ## 5. Fast-Track & Quick-Fix Guidelines (Lean Mode)
36
-
37
- - For trivial bug fixes, typo corrections, or single-file non-architectural changes, developers and agents may use the **Quick-Fix Fast-Track**:
38
- - Consolidate Discovery, Define, and Spec into an inline fix brief (`devflow/runs/{fix-id}/fix-brief.md`).
39
- - Jump directly to `40-execute` (with reproduction unit test) ➔ `50-verify` ➔ `70-release`.
40
- - Avoid creating heavy redundant documentation for changes under 20 lines of code while preserving Empirical Proof and Unit Test requirements.
95
+ - `open`: Confirmed issue waiting to be fixed.
96
+ - `fixed`: Repaired in code, pending QA re-verification.
97
+ - `closed`: Verified by QA as completely resolved without regressions.
98
+ - `accepted`: Formally waived with recorded user justification.
99
+ - **P0/P1 Blockers**: Any P0 or P1 finding in `open` or `fixed` status unconditionally blocks `/complete` and `70-release`.
100
+
101
+ ---
102
+
103
+ ## 7. Error Handling & Exit Codes
104
+
105
+ - Use structured `try / catch` blocks around filesystem and process operations.
106
+ - Create meaningful, actionable error messages for the user (explain what failed and how to resolve it).
107
+ - Exit cleanly with standard POSIX process exit codes:
108
+ - `0`: Successful execution.
109
+ - `1`: Validation failure, missing arguments, or runtime error.
110
+
111
+ ---
112
+
113
+ ## 8. Comments & Documentation Discipline
114
+
115
+ Write code that explains itself; comment only what the code cannot say:
116
+
117
+ - Comment the **why**, not the **what**. Avoid comments that simply rephrase the line of code.
118
+ - Avoid noisy banner blocks, section divider lines (`// ====================`), or obvious step narrations.
119
+ - Comments earn their place when documenting:
120
+ - Non-obvious architectural decisions.
121
+ - Upstream compatibility quirks or workarounds.
122
+ - Edge-case handling rationale.
123
+ - Keep JSDoc minimal and useful: provide a one-line summary and document non-obvious parameters.
124
+
125
+ ---
126
+
127
+ ## 9. Writing & Language Conventions
128
+
129
+ - **Default Communication & Artifacts**: Thai (`th`) for all generated markdown stage artifacts (`current-feature.md`, `00-discover.md`, etc.), explanations, and user interactions.
130
+ - **Code & Identifiers**: English for all source code, variable names, file paths, CLI flags, and commit messages.
131
+ - **Typography**: Do not use em dashes (`—`) in AI-generated commit messages or technical summaries; use standard hyphens (`-`) or colons.
@@ -0,0 +1,3 @@
1
+ # Current Feature
2
+
3
+ _Nothing in progress. Run /feature, /fix, or /rollback to start._
@@ -1,8 +1,8 @@
1
1
  # Current DevFlow Run Status
2
2
 
3
- - **Active Discovery ID**: None
4
- - **Active Running ID**: None
5
- - **Current Stage**: Idle (Ready for next run)
6
- - **Living Spec**: None
7
- - **Last Completed Run**: RUN-019-sync-upstream-status-cli-and-project-detection (2026-08-20)
8
- - **Last Updated**: 2026-08-20
3
+ - **Active Discovery ID**: `None`
4
+ - **Active Running ID**: `None`
5
+ - **Current Stage**: `Idle (Ready for new /feature, /fix, /00-discover, or /10-define)`
6
+ - **Living Spec**: `None`
7
+ - **Last Completed Run**: `None`
8
+ - **Last Updated**: `None`
@@ -1,13 +1,13 @@
1
- # Audit Findings Ledger
1
+ # Findings Ledger
2
2
 
3
- Summary of audit findings, quality checks, security bugs, and debt tracked across DevFlow runs.
3
+ > **Generated Ledger File.** Tracks quality findings, security vulnerabilities, regression risks, and architectural debt raised during review and QA passes.
4
+ >
5
+ > **Finding Format**:
6
+ > `### <ID> [<SEVERITY>] <STATUS> - <Title>`
7
+ > - **Severities**: `P0` (Critical Blocker), `P1` (High Blocker), `P2` (Medium), `P3` (Low / Polish)
8
+ > - **Statuses**: `unverified`, `open`, `fixed`, `closed`, `accepted`, `invalid`
9
+ > - **Release Gate Rule**: Any `P0` or `P1` finding in `open` or `fixed` status unconditionally blocks `/complete` and `70-release`.
4
10
 
5
- ## Open Findings
11
+ ---
6
12
 
7
- | ID | Severity | Category | Title | Discovered Run | Status |
8
- | :- | :--- | :--- | :--- | :--- | :--- |
9
-
10
- ## Resolved Findings
11
-
12
- | ID | Severity | Category | Title | Resolved Run | Date |
13
- | :- | :--- | :--- | :--- | :--- | :--- |
13
+ _No active findings recorded. QA and audit passes append findings here as they are discovered._
@@ -1,41 +1,27 @@
1
- # Project Overview
1
+ # Project Overview & Source of Truth
2
2
 
3
- Primary source of truth for the project context, architecture, tech stack, and active DevFlow stage status.
3
+ > Living context artifact automatically synchronized with codebase reality and DevFlow delivery history.
4
+ > Run `/onboard` (for new projects) or `/adopt` (for existing codebases) to populate this file.
4
5
 
5
- ## Application Summary
6
+ ---
6
7
 
7
- - **Project Name**: Nexus-DevFlow
8
- - **Goal**: Agent-ready DevFlow 2.0 workflow layer framework with .agents and .claude adapter tooling, living context synchronization, and automated AI Blueprint upstream monitoring.
9
- - **Target Audience**: AI coding assistants (Google Antigravity, OpenAI Codex, Claude Code, Cursor) and software development teams building resilient agentic delivery workflows.
8
+ ## 1. Project Purpose & Target Users
9
+ - Describe what this application does, who it is for, and the problems it solves.
10
10
 
11
- ## Tech Stack
11
+ ## 2. Architecture & Directory Layout
12
+ - Visual directory layout and overview of core modules.
12
13
 
13
- - **Core & Runtime**: Node.js (>=18.17 / Node 22), TypeScript 5.7, Node Next ESM (`"type": "module"`)
14
- - **Execution & Tooling**: `tsx` (TypeScript execute/test runner), `tsc` (Typecheck & compilation)
15
- - **Installer Package**: `@jakkrichm/create-nexus-devflow` (with `dist/` compilation pipeline)
16
- - **Documentation**: VitePress / Astro website under `website/` (GitHub Pages)
17
- - **Upstream Integration**: Automated GitHub Actions monitor (`.github/workflows/check-upstream.yml`) tracking `aiblueprinthq/ai-blueprint`
18
- - **Verification Gate**: `npm run check` (TypeScript typechecking, static contract validation, TF-IDF skill routing evaluations, installer unit tests, packed smoke tests)
14
+ ## 3. Technology Stack & Key Tooling
15
+ - Frameworks, languages, databases, ORMs, package managers, and runtime environment.
19
16
 
20
- ## Architecture & Key Modules
17
+ ## 4. Concrete Data Models & Entities
18
+ - Core entities, database schemas, API interfaces, and types.
21
19
 
22
- - **Dual-Track Delivery Engine**:
23
- - **Fast-Track (Blueprint Mode 4 Steps)**: High-velocity spec loop (`/feature` or `/fix` ➔ `/implement` ➔ `/check` ➔ `/complete`) with Single Living Spec (`spec.md`).
24
- - **Deep-Track (Architect Mode — 8 Steps)**: Full-lifecycle delivery pipeline (`00-discover` ➔ `10-define` ➔ `20-spec` ➔ `30-plan` ➔ `40-execute` ➔ `50-verify` ➔ `60-report` ➔ `70-release`).
25
- - **`packages/create-nexus-devflow/`**: Distribution CLI package with TypeScript compilation (`dist/bin/create-nexus-devflow.js`), atomic update engine (`lib/update.ts`), and prepack template bundler (`scripts/prepare-template.ts`).
26
- - **`.agents/skills/` & `.claude/skills/`**: 80 synchronized AI agent skills covering Dual-Track workflows and companion tools (`devflow`, `idea`, `report-html`, `try`, `doctor`, `overview`, `sync-upstream`, `ci`, `rollback`).
27
- - **`scripts/`**: Maintainer infrastructure (`check-devflow.ts`, `validate-framework.ts`, `smoke-package.ts`, `upstream-monitor.ts`, `evals/routing.ts`).
28
- - **`devflow/`**: Workspace state, discoveries, runs, ideas, context, reference contracts, and master release history ledger.
20
+ ## 5. Shipped Capabilities & Milestones
21
+ - Verified features and components currently functional in the codebase.
29
22
 
30
- ## Shipped Capabilities (Recent Milestones)
23
+ ## 6. Verified Commands & Developer Workflow
24
+ - Dev Server, Build, Test, Lint, and Verify commands.
31
25
 
32
- - **`RUN-017` (2026-08-20)**: Separate Fast-Track skills (`/feature`, `/fix`) and rename Deep-Track stage 40 to `40-execute` to eliminate command collisions.
33
- - **`RUN-016` (2026-08-20)**: Quick Idea Capture and AI Feasibility Assessment (`/idea`) with centralized Idea Inbox (`devflow/ideas.md`).
34
- - **`RUN-015` (2026-08-20)**: Dual-Track Architecture (Fast-Track 4 Steps & Deep-Track 8 Steps) + Living Spec (`spec.md`) + Standalone HTML Reporting Policy.
35
- - **`RUN-014` (2026-08-20)**: Upgraded DevFlow to full TypeScript architecture (`tsconfig.json`, `tsx`, `tsc`, `dist/`), migrated AI Blueprint Upstream Monitor workflow to DevFlow, and established multi-lane verification matrix.
36
-
37
- ## DevFlow Workspace Status
38
-
39
- - **Active Discovery ID**: None (Idle)
40
- - **Active Running ID**: `RUN-018-update-documentation-and-guides`
41
- - **Last Completed Run**: `RUN-017-split-spec-and-rename-40-execute`
26
+ ## 7. Known Architectural Focus Areas
27
+ - Upcoming priorities, refactoring targets, or known technical considerations.
@@ -1,34 +1,19 @@
1
- # Nexus-DevFlow Master Release History Ledger
1
+ # Master Release History Ledger
2
2
 
3
- This master ledger tracks all released delivery runs, milestones, and rollbacks in chronological order. Each entry is recorded during the `70-release` stage and links to its exact Git commit hash, release tag, and generated delivery artifacts.
3
+ This master ledger tracks all released delivery runs, milestones, and rollbacks in chronological order. Each entry is recorded during `/complete` or `70-release` and links to its exact Git commit hash, release status, category, and archived delivery artifacts.
4
4
 
5
5
  ---
6
6
 
7
- ## 📜 Release Log
7
+ ## 📜 Master Release Log
8
8
 
9
- | Release Date | Running ID | Title | Git Tag / Commit | Release Status | Report Link |
10
- | :--- | :--- | :--- | :--- | :--- | :--- |
11
- | 2026-08-20 | `RUN-019` | ซิงก์ฟีเจอร์ Status CLI, Project Detection, Unit Tests และ Upstream Baseline จาก AI Blueprint v0.9.1 | `HEAD` | `Released` | [`60-report.md`](../runs/RUN-019-sync-upstream-status-cli-and-project-detection/60-report.md) |
12
- | 2026-08-20 | `RUN-018` | อัปเดตคู่มือ เอกสาร และ Website Documentation ให้เป็นปัจจุบันและครอบคลุม Dual-Track Model ล่าสุด | `HEAD` | `Released` | [`spec.md`](../runs/RUN-018-update-documentation-and-guides/spec.md) |
13
- | 2026-08-20 | `RUN-017` | แยกคำสั่ง Fast-Track เป็น `/feature`, `/fix` และเปลี่ยน Deep-Track สเตจ 40 เป็น `40-execute` | `HEAD` | `Released` | [`spec.md`](../runs/RUN-017-split-spec-and-rename-40-execute/spec.md) |
14
- | 2026-08-20 | `RUN-016` | ระบบบันทึกไอเดียพร้อม AI วิเคราะห์ความเป็นไปได้ (`/idea`) และเชื่อมต่อกับ Status Backlog | `HEAD` | `Released` | [`spec.md`](../runs/RUN-016-idea-capture-inbox-and-status-integration/spec.md) |
15
- | 2026-08-20 | `RUN-015` | Dual-Track Architecture (Fast-Track 4 Steps & Deep-Track 8 Steps) + Living Spec (`spec.md`) + Standalone HTML Report Policy | `HEAD` | `Released` | [`spec.md`](../runs/RUN-015-fast-track-and-living-blueprint/spec.md) |
16
- | 2026-08-20 | `RUN-014` | ยกระดับสถาปัตยกรรม DevFlow สู่ TypeScript และย้ายระบบ Check AI Blueprint Upstream Monitor | `HEAD` | `Released` | [`60-report.md`](../runs/RUN-014-typescript-migration-and-upstream-monitor-for-devflow/60-report.md) |
17
- | 2026-08-20 | `RUN-013` | เพิ่ม Skill `/overview` และระบบ Living Context Sync สำหรับ Nexus-DevFlow | `e30375b` | `Released` | [`60-report.md`](../runs/RUN-013-add-overview-and-context-sync-skill/60-report.md) |
18
- | 2026-08-18 | `RUN-008` | ปรับปรุงโครงสร้าง Nexus-DevFlow ให้ Lean & Clean ยุบรวม Skills บริหารจัดการ History และ Safe Rollback | `be713ea` | `Released` | [`60-report.md`](../runs/RUN-008-lean-and-clean-devflow-optimization/60-report.md) |
19
- | 2026-08-18 | `RUN-007` | ยกระดับระบบ Nexus-DevFlow ด้วยวินัยและกลไกสำคัญจาก Blueprint | `e79528b` | `Released` | [`60-report.md`](../runs/RUN-007-integrate-blueprint-skills-enhancements/60-report.md) |
20
- | 2026-08-18 | `RUN-006` | Standardize Command Naming and Provider Invocation | `HEAD` | `Released` | [`60-report.md`](../runs/RUN-006-standardize-command-naming-and-provider-invocation/60-report.md) |
21
- | 2026-08-18 | `RUN-005` | Add DevFlow Prefix to Skill Descriptions | `HEAD` | `Released` | [`60-report.md`](../runs/RUN-005-add-devflow-prefix-to-skill-descriptions/60-report.md) |
22
- | 2026-08-18 | `RUN-004` | Add Autopilot Skill | `HEAD` | `Released` | [`60-report.md`](../runs/RUN-004-add-autopilot-skill/60-report.md) |
23
- | 2026-08-18 | `RUN-003` | Add Try, Rollback, CI, Brief Skills | `HEAD` | `Released` | [`60-report.md`](../runs/RUN-003-add-try-rollback-ci-brief-skills/60-report.md) |
24
- | 2026-08-18 | `RUN-002` | Add Onboard, Adopt, Doctor Skills | `HEAD` | `Released` | [`60-report.md`](../runs/RUN-002-add-onboard-adopt-doctor-skills/60-report.md) |
25
- | 2026-08-18 | `RUN-001` | Align DevFlow with Blueprint Architecture | `HEAD` | `Released` | [`60-report.md`](../runs/RUN-001-align-devflow-blueprint/60-report.md) |
9
+ | Completed Date | Run ID | Category | Title | Git Commit | Status | Archive Link |
10
+ | :--- | :--- | :--- | :--- | :--- | :--- | :--- |
11
+ | _No shipped runs yet_ | - | - | Run `/feature` or `/00-discover` to start your first delivery run | - | - | - |
26
12
 
27
13
  ---
28
14
 
29
- ## 🗄️ Run Lifecycle States
15
+ ## 🗄️ History Categories (The Core 3 Model)
30
16
 
31
- - **Active**: Currently executing in `devflow/runs/{running-id}-{slug}/` (tracked in `devflow/context/current-stage.md`).
32
- - **Released**: Finished `70-release`, merged, and logged in this master ledger.
33
- - **Archived**: Older historical runs moved to `devflow/history/archived-runs/` to maintain a clean workspace.
34
- - **Rolled Back**: Features safely reversed via `/rollback` with documented post-reversal verification.
17
+ - **`features/`**: New user-facing features, enhancements, architecture migrations, refactoring, and tooling/infra.
18
+ - **`fixes/`**: Bug fixes, hotfixes, regressions, security patches, and performance optimizations.
19
+ - **`rollbacks/`**: Safe feature reversal and rollback execution records.
@@ -1,15 +1,19 @@
1
- # 💡 DevFlow Idea Inbox & Backlog
1
+ # 🔮 Centralized Idea Inbox & Backlog
2
2
 
3
- บันทึกไอเดียที่รอดำเนินการ พร้อมบทวิเคราะห์ความเป็นไปได้เบื้องต้นจาก AI (บันทึกด้วยคำสั่ง `/idea "<text>"`)
3
+ > Record, analyze, and prioritize feature ideas before starting active delivery.
4
+ > Use the `/idea` command to quickly capture and evaluate new ideas.
4
5
 
5
6
  ---
6
7
 
7
- ## 📌 Pending Ideas
8
+ ## 💡 Active Idea Inbox
8
9
 
9
- *(ยังไม่มีไอเดียค้างอยู่ เพิ่มไอเดียใหม่ได้ด้วยคำสั่ง `/idea "<ข้อความไอเดีย>" หรือ `$idea "<text>"`)*
10
+ | ID | Title | Value (1-5) | Feasibility (1-5) | Status | Notes |
11
+ | :--- | :--- | :---: | :---: | :--- | :--- |
12
+ | _No pending ideas_ | - | - | - | Use `/idea <your idea>` to capture | - |
10
13
 
11
14
  ---
12
15
 
13
16
  ## 📦 Archived / Shipped Ideas
14
17
 
15
- *(ไอเดียที่ถูกหยิบไปทำด้วย `/feature IDEA-xxx`, `/fix IDEA-xxx` หรือ `/00-discover IDEA-xxx` จะถูกย้ายมาบันทึกที่นี่)*
18
+ | ID | Title | Shipped In | Completed Date |
19
+ | :--- | :--- | :--- | :--- |
@@ -1,39 +1,39 @@
1
- # DevFlow 2.0 Running ID & Workspace Contract Reference
2
-
3
- ## DevFlow Workspace Directory Structure
4
-
5
- All DevFlow framework context, active runs, templates, and history are consolidated under `devflow/`:
6
-
7
- - **Context (`devflow/context/`)**:
8
- - `project-overview.md` - Primary source of truth for project architecture and stack.
9
- - `coding-standards.md` - Development, code quality, and testing standards.
10
- - `ai-interaction.md` - AI agent interaction rules and Thai artifact defaults.
11
- - `current-stage.md` - Active stage run state.
12
- - `findings.md` - Open and resolved audit findings ledger.
13
-
14
- - **Active Delivery Runs (`devflow/runs/{RUNNING_ID}/`)**:
15
- - `10-define.md`
16
- - `20-spec.md`
17
- - `30-plan.md`
18
- - `40-execute.md`
19
- - `50-verify.md`
20
- - `60-report.md` (and `60-report.html`)
21
- - `70-release.md`
22
-
23
- - **Active Discoveries (`devflow/discoveries/{DISCOVERY_ID}-{slug}/`)**:
24
- - `00-discover.md`
25
-
26
- - **History Archive (`devflow/history/`)**:
27
- - `discoveries/` - Completed discovery archives.
28
- - `runs/` - Completed delivery run archives.
29
- - `reports/` - Completed stage reports.
30
-
31
- - **Reference Contracts (`devflow/reference/`)**:
32
- - `running-id-contract.md`
33
- - `stage-lifecycle-contract.md`
34
-
35
- ## Mainline Lifecycle Rules
36
-
37
- 1. Mainline stages move linearly: `/00-Discover` `/10-Define` `/20-Spec` `/30-Plan` `/40-Execute` `/50-Verify` ➔ `/60-Report` ➔ `/70-Release`.
38
- 2. Companion commands (`Goal`, `Brainstorm`, `Research`, `Debug`, `PRD`, `Issue-Triage`, `Security-Review`, `Check-For-Updates`, `Help`) provide supporting context without replacing mainline stage numbers.
39
- 3. Every stage produces markdown-first evidence in `devflow/runs/` before passing verification gates.
1
+ # DevFlow 2.0 Running ID & 3-Pillars Workspace Contract
2
+
3
+ ## The 3-Pillars Workspace Architecture
4
+
5
+ All DevFlow framework assets are organized into three clean pillars representing Future, Present, and Past:
6
+
7
+ ```text
8
+ devflow/
9
+ ├── 🔮 ideas.md # [1. Future / Backlog] Centralized Idea Inbox with AI scoring
10
+
11
+ ├── ⚡ context/ # [2. Present / Active] Living Source of Truth & Active Work
12
+ │ ├── project-overview.md # Primary source of truth for project architecture and tech stack
13
+ │ ├── coding-standards.md # Engineering, code quality, TDD, and testing standards
14
+ │ ├── ai-interaction.md # AI agent interaction rules, dual-track flow, and Thai defaults
15
+ │ ├── findings.md # Open and resolved audit findings ledger (P0-P3)
16
+ │ ├── current-stage.md # Active state pointer and run tracker
17
+ │ ├── current-feature.md # Fast-Track Single Living Spec (Active work / stub when idle)
18
+ │ └── current-run/ # Deep-Track Active Run folder (Temporary during 10-70 execution)
19
+
20
+ ├── 📦 history/ # [3. Past / Completed] Permanent Delivery & Release Archives
21
+ │ ├── features/ # Completed features, architecture migrations, tooling (xxx-slug.md or xxx-slug/)
22
+ │ ├── fixes/ # Completed bug fixes, hotfixes, security patches (xxx-slug.md)
23
+ │ ├── rollbacks/ # Completed feature reversals (YYYY-MM-DD-xxx-slug.md)
24
+ │ └── HISTORY.md # Master release ledger summary table
25
+
26
+ └── 🔍 discoveries/ # Pre-delivery discovery records (DISC-YYYYMMDD-NNN-slug/00-discover.md)
27
+ ```
28
+
29
+ ## Running ID Naming Convention
30
+
31
+ - **Format**: `xxx-slug` (e.g. `001-setup-auth`, `021-categorized-history-and-clean-living-spec-architecture`)
32
+ - **Prefix Removal**: The legacy `RUN-` prefix is discontinued in favor of clean 3-digit sequential numbering.
33
+ - **Git Branch Standard**: `feature/{xxx-slug}` or `fix/{xxx-slug}`.
34
+
35
+ ## Single Active Run Rule (One Thing at a Time)
36
+
37
+ 1. Only **one active run** is permitted at any given time across both Fast-Track and Deep-Track.
38
+ 2. Before opening a new run (`/feature`, `/fix`, or `10-define`), the AI checks `current-stage.md` and `current-feature.md`.
39
+ 3. If an active run is in progress, the AI **blocks** starting a new task and requires closing the active run via `/complete` or `70-release` (or explicitly cancelling/rolling back).
File without changes