@thierrynakoa/fire-flow 10.0.0 → 12.2.0
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.
- package/.claude-plugin/plugin.json +9 -9
- package/ARCHITECTURE-DIAGRAM.md +7 -4
- package/COMMAND-REFERENCE.md +33 -13
- package/DOMINION-FLOW-OVERVIEW.md +581 -421
- package/QUICK-START.md +3 -3
- package/README.md +102 -45
- package/TROUBLESHOOTING.md +264 -264
- package/agents/fire-executor.md +200 -116
- package/agents/fire-fact-checker.md +276 -276
- package/agents/fire-phoenix-analyst.md +394 -0
- package/agents/fire-planner.md +145 -53
- package/agents/fire-project-researcher.md +155 -155
- package/agents/fire-research-synthesizer.md +166 -166
- package/agents/fire-researcher.md +144 -59
- package/agents/fire-roadmapper.md +215 -203
- package/agents/fire-verifier.md +247 -65
- package/agents/fire-vision-architect.md +381 -0
- package/commands/fire-0-orient.md +476 -476
- package/commands/fire-1a-new.md +216 -0
- package/commands/fire-1b-research.md +210 -0
- package/commands/fire-1c-setup.md +254 -0
- package/commands/{fire-1a-discuss.md → fire-1d-discuss.md} +35 -7
- package/commands/fire-3-execute.md +55 -2
- package/commands/fire-4-verify.md +61 -0
- package/commands/fire-5-handoff.md +2 -2
- package/commands/fire-6-resume.md +37 -2
- package/commands/fire-add-new-skill.md +2 -2
- package/commands/fire-autonomous.md +20 -3
- package/commands/fire-brainstorm.md +1 -1
- package/commands/fire-complete-milestone.md +2 -2
- package/commands/fire-cost.md +183 -0
- package/commands/fire-dashboard.md +2 -2
- package/commands/fire-debug.md +663 -663
- package/commands/fire-loop-resume.md +2 -2
- package/commands/fire-loop-stop.md +1 -1
- package/commands/fire-loop.md +1168 -1168
- package/commands/fire-map-codebase.md +3 -3
- package/commands/fire-new-milestone.md +356 -356
- package/commands/fire-phoenix.md +603 -0
- package/commands/fire-reflect.md +235 -235
- package/commands/fire-research.md +246 -246
- package/commands/fire-search.md +1 -1
- package/commands/fire-skills-diff.md +3 -3
- package/commands/fire-skills-history.md +3 -3
- package/commands/fire-skills-rollback.md +7 -7
- package/commands/fire-skills-sync.md +5 -5
- package/commands/fire-test.md +9 -9
- package/commands/fire-todos.md +1 -1
- package/commands/fire-update.md +5 -5
- package/hooks/hooks.json +16 -16
- package/hooks/run-hook.sh +8 -8
- package/hooks/run-session-end.sh +7 -7
- package/hooks/session-end.sh +90 -90
- package/hooks/session-start.sh +1 -1
- package/package.json +2 -2
- package/plugin.json +7 -7
- package/references/metrics-and-trends.md +1 -1
- package/skills-library/SKILLS-INDEX.md +588 -588
- package/skills-library/_general/methodology/AUTONOMOUS_ORCHESTRATION.md +182 -0
- package/skills-library/_general/methodology/BACKWARD_PLANNING_INTERVIEW.md +307 -0
- package/skills-library/_general/methodology/CIRCUIT_BREAKER_INTELLIGENCE.md +163 -0
- package/skills-library/_general/methodology/CONTEXT_ROTATION.md +151 -0
- package/skills-library/_general/methodology/DEAD_ENDS_SHELF.md +188 -0
- package/skills-library/_general/methodology/DESIGN_PHILOSOPHY_ENFORCEMENT.md +152 -0
- package/skills-library/_general/methodology/INTERNAL_CONSISTENCY_AUDIT.md +212 -0
- package/skills-library/_general/methodology/LIVE_BREADCRUMB_PROTOCOL.md +242 -0
- package/skills-library/_general/methodology/PHOENIX_REBUILD_METHODOLOGY.md +251 -0
- package/skills-library/_general/methodology/QUALITY_GATES_AND_VERIFICATION.md +157 -0
- package/skills-library/_general/methodology/RELIABILITY_PREDICTION.md +104 -0
- package/skills-library/_general/methodology/REQUIREMENTS_DECOMPOSITION.md +155 -0
- package/skills-library/_general/methodology/SELF_TESTING_FEEDBACK_LOOP.md +143 -0
- package/skills-library/_general/methodology/STACK_COMPATIBILITY_MATRIX.md +178 -0
- package/skills-library/_general/methodology/TIERED_CONTEXT_ARCHITECTURE.md +118 -0
- package/skills-library/_general/methodology/ZERO_FRICTION_CLI_SETUP.md +312 -0
- package/skills-library/_general/methodology/autonomous-multi-phase-build.md +133 -0
- package/skills-library/_general/methodology/claude-md-archival.md +280 -0
- package/skills-library/_general/methodology/debug-swarm-researcher-escape-hatch.md +240 -240
- package/skills-library/_general/methodology/git-worktrees-parallel.md +232 -0
- package/skills-library/_general/methodology/llm-judge-memory-crud.md +241 -0
- package/skills-library/_general/methodology/multi-project-autonomous-build.md +360 -0
- package/skills-library/_general/methodology/shell-autonomous-loop-fixplan.md +238 -238
- package/skills-library/_general/patterns-standards/GOF_DESIGN_PATTERNS_FOR_AI_AGENTS.md +358 -0
- package/skills-library/methodology/BREATH_BASED_PARALLEL_EXECUTION.md +1 -1
- package/skills-library/methodology/RESEARCH_BACKED_WORKFLOW_UPGRADE.md +1 -1
- package/skills-library/methodology/SABBATH_REST_PATTERN.md +1 -1
- package/templates/ASSUMPTIONS.md +1 -1
- package/templates/BLOCKERS.md +1 -1
- package/templates/DECISION_LOG.md +1 -1
- package/templates/phase-prompt.md +1 -1
- package/templates/phoenix-comparison.md +80 -0
- package/version.json +2 -2
- package/workflows/handoff-session.md +1 -1
- package/workflows/new-project.md +2 -2
- package/commands/fire-1-new.md +0 -281
package/QUICK-START.md
CHANGED
|
@@ -43,7 +43,7 @@ Dominion Flow uses 6 numbered commands that take your project from start to fini
|
|
|
43
43
|
## Step 1: Start a New Project
|
|
44
44
|
|
|
45
45
|
```bash
|
|
46
|
-
/fire-
|
|
46
|
+
/fire-1a-new
|
|
47
47
|
```
|
|
48
48
|
|
|
49
49
|
This command:
|
|
@@ -311,7 +311,7 @@ When you solve a new problem, add it to the library:
|
|
|
311
311
|
|
|
312
312
|
### Starting Fresh
|
|
313
313
|
```bash
|
|
314
|
-
/fire-
|
|
314
|
+
/fire-1a-new
|
|
315
315
|
/fire-2-plan 1
|
|
316
316
|
/fire-3-execute 1
|
|
317
317
|
/fire-4-verify 1
|
|
@@ -320,7 +320,7 @@ When you solve a new problem, add it to the library:
|
|
|
320
320
|
|
|
321
321
|
### Full Autopilot (after project init)
|
|
322
322
|
```bash
|
|
323
|
-
/fire-
|
|
323
|
+
/fire-1a-new
|
|
324
324
|
/fire-autonomous
|
|
325
325
|
```
|
|
326
326
|
|
package/README.md
CHANGED
|
@@ -1,20 +1,51 @@
|
|
|
1
|
-
# Dominion Flow
|
|
1
|
+
# Fire Flow (Dominion Flow) v12.2
|
|
2
2
|
|
|
3
|
-
**
|
|
3
|
+
**The most comprehensive orchestration platform for Claude Code.**
|
|
4
4
|
|
|
5
|
-
|
|
5
|
+
Fire Flow gives Claude a complete, structured way to take your project from idea to finished code — with built-in quality checks, session memory, parallel execution, and a library of 478+ proven patterns. Think of it as a project management system that lives inside Claude Code.
|
|
6
|
+
|
|
7
|
+
---
|
|
8
|
+
|
|
9
|
+
## NEW in v12.2: Phoenix Rebuild
|
|
10
|
+
|
|
11
|
+
**Every AI coding tool creates code. Fire Flow is the first that can take a messy "vibe coded" project and rebuild it clean.**
|
|
12
|
+
|
|
13
|
+
```bash
|
|
14
|
+
/fire-phoenix --source ./my-messy-app
|
|
15
|
+
```
|
|
16
|
+
|
|
17
|
+
Phoenix reverse-engineers what your code was *trying* to do, asks you clarifying questions, then rebuilds the entire project from scratch in a new folder — production-ready, properly architected, fully tested.
|
|
18
|
+
|
|
19
|
+
**6-phase autonomous pipeline:**
|
|
20
|
+
```
|
|
21
|
+
AUTOPSY → INTENT → CLARIFY → VISION → REBUILD → COMPARISON
|
|
22
|
+
```
|
|
23
|
+
|
|
24
|
+
- Extracts developer INTENT from messy code (not just what it does, but what it was *trying* to do)
|
|
25
|
+
- Classifies every feature: intended, accidental, workaround, or abandoned
|
|
26
|
+
- Maps anti-patterns to clean replacements (12 common vibe-coder patterns)
|
|
27
|
+
- Preserves edge cases and business rules (the important stuff)
|
|
28
|
+
- Source folder is never modified — rebuild goes to a new target folder
|
|
29
|
+
- Phoenix Score (PX-1 to PX-5) verifies feature parity, edge case coverage, and architecture improvement
|
|
30
|
+
|
|
31
|
+
**Dry-run mode** analyzes without rebuilding — useful for auditing any codebase:
|
|
32
|
+
```bash
|
|
33
|
+
/fire-phoenix --source ./app --dry-run
|
|
34
|
+
```
|
|
6
35
|
|
|
7
36
|
---
|
|
8
37
|
|
|
9
38
|
## What Does It Do?
|
|
10
39
|
|
|
11
|
-
When you start a new project, Claude normally has no memory between sessions, no standard process, and no way to verify its own work.
|
|
40
|
+
When you start a new project, Claude normally has no memory between sessions, no standard process, and no way to verify its own work. Fire Flow fixes all of that:
|
|
12
41
|
|
|
13
42
|
- **Structured workflow** — A numbered pipeline (Plan → Execute → Verify → Handoff) so nothing gets skipped
|
|
14
43
|
- **Session memory** — Claude picks up exactly where it left off, every time
|
|
15
44
|
- **Parallel execution** — Multiple tasks run at the same time, safely, so work gets done faster
|
|
16
|
-
- **Built-in quality gates** —
|
|
17
|
-
- **Skills library** —
|
|
45
|
+
- **Built-in quality gates** — 70-point verification checklist with tiered gates (fast → must-haves → comprehensive)
|
|
46
|
+
- **Skills library** — 478+ proven code patterns Claude can reuse instead of reinventing every time
|
|
47
|
+
- **Phoenix Rebuild** — Take any messy codebase and rebuild it clean from extracted intent
|
|
48
|
+
- **Research-backed methodology** — Circuit breaker intelligence, kill conditions, stuck-state classification, and more
|
|
18
49
|
|
|
19
50
|
---
|
|
20
51
|
|
|
@@ -24,39 +55,26 @@ This plugin is for anyone using Claude Code who wants:
|
|
|
24
55
|
- Consistent, repeatable results on complex projects
|
|
25
56
|
- Claude to remember what it was doing between sessions
|
|
26
57
|
- Code that gets reviewed and verified, not just written
|
|
58
|
+
- A messy "vibe coded" project cleaned up and rebuilt properly
|
|
27
59
|
|
|
28
60
|
**No prior experience with orchestration or AI agents required.**
|
|
29
61
|
|
|
62
|
+
---
|
|
30
63
|
|
|
64
|
+
## Quick Install
|
|
31
65
|
|
|
32
|
-
|
|
33
|
-
|
|
34
|
-
Are you using Claude Code to build projects but feeling limited by session-hopping, lack of structure, or the need to constantly re-explain your requirements? Dominion Flow is a project management and orchestration plugin built to transform Claude from a simple assistant into a structured, persistent, and highly capable autonomous development agent.
|
|
35
|
-
What is Dominion Flow?
|
|
36
|
-
|
|
37
|
-
Dominion Flow acts as a professional-grade "operating system" inside your terminal. It introduces a formal, repeatable pipeline—Plan → Execute → Verify → Handoff—that ensures your project moves from an idea to working, verified code without skipping critical steps.
|
|
38
|
-
Why Use It?
|
|
39
|
-
|
|
40
|
-
Persistent Memory: Through integrated vector database support (Qdrant), Dominion Flow allows Claude to remember your codebase, past decisions, and coding patterns across different sessions.
|
|
41
|
-
|
|
42
|
-
Structured Pipeline: Stop guessing where to start. The platform uses a clear, 39-command framework to guide you through every phase of development.
|
|
43
|
-
|
|
44
|
-
Quality & Verification: Built-in quality gates and automated testing (including Playwright E2E testing) ensure that code isn't just written—it's verified.
|
|
45
|
-
|
|
46
|
-
Autonomous Capabilities: Need to go hands-off? Features like /fire-autonomous allow Claude to plan, code, and verify entire project phases on its own.
|
|
47
|
-
|
|
48
|
-
Reusable Skills Library: Stop reinventing the wheel. Dominion Flow includes a library of proven, reusable patterns for authentication, APIs, payments, and more, which Claude can automatically learn and store as you build.
|
|
49
|
-
|
|
50
|
-
Who Is It For?
|
|
66
|
+
One command does it all:
|
|
51
67
|
|
|
52
|
-
|
|
53
|
-
|
|
68
|
+
```bash
|
|
69
|
+
npx @thierrynakoa/fire-flow
|
|
70
|
+
```
|
|
54
71
|
|
|
55
|
-
|
|
56
|
-
Bash
|
|
72
|
+
Or clone the repository manually:
|
|
57
73
|
|
|
74
|
+
```bash
|
|
58
75
|
git clone https://github.com/ThierryN/fire-flow.git
|
|
59
76
|
claude install-plugin ./fire-flow
|
|
77
|
+
```
|
|
60
78
|
|
|
61
79
|
For advanced users, the repository also supports optional "Power Features" like Docker-integrated memory (Qdrant) and local embeddings (Ollama) to make the agent even more powerful.
|
|
62
80
|
|
|
@@ -68,11 +86,27 @@ For advanced users, the repository also supports optional "Power Features" like
|
|
|
68
86
|
|
|
69
87
|
**Prerequisite:** You need [Claude Code](https://claude.ai/download) installed first. If you don't have it yet, download and install it, then come back here.
|
|
70
88
|
|
|
71
|
-
|
|
89
|
+
### Method A — One Command (Recommended)
|
|
90
|
+
|
|
91
|
+
```bash
|
|
92
|
+
npx @thierrynakoa/fire-flow
|
|
93
|
+
```
|
|
94
|
+
|
|
95
|
+
That's it. The installer copies everything to `~/.claude/plugins/fire-flow/` automatically.
|
|
96
|
+
|
|
97
|
+
To update later:
|
|
98
|
+
```bash
|
|
99
|
+
npx @thierrynakoa/fire-flow --update
|
|
100
|
+
```
|
|
101
|
+
|
|
102
|
+
To uninstall:
|
|
103
|
+
```bash
|
|
104
|
+
npx @thierrynakoa/fire-flow --uninstall
|
|
105
|
+
```
|
|
72
106
|
|
|
73
107
|
---
|
|
74
108
|
|
|
75
|
-
### Method
|
|
109
|
+
### Method B — Git Clone
|
|
76
110
|
|
|
77
111
|
1. Open your terminal and clone the repo:
|
|
78
112
|
```bash
|
|
@@ -90,7 +124,7 @@ Choose either Method A (recommended) or Method B:
|
|
|
90
124
|
|
|
91
125
|
---
|
|
92
126
|
|
|
93
|
-
### Method
|
|
127
|
+
### Method C — Download ZIP (No Git or npm Required)
|
|
94
128
|
|
|
95
129
|
1. Go to [github.com/ThierryN/fire-flow](https://github.com/ThierryN/fire-flow)
|
|
96
130
|
2. Click the green **"Code"** button → **"Download ZIP"**
|
|
@@ -262,14 +296,14 @@ Claude will query Qdrant directly and confirm it is live.
|
|
|
262
296
|
Start a new project with one command:
|
|
263
297
|
|
|
264
298
|
```bash
|
|
265
|
-
/fire-
|
|
299
|
+
/fire-1a-new
|
|
266
300
|
```
|
|
267
301
|
|
|
268
302
|
Claude will ask you a few simple questions about your project, then set everything up automatically. After that, the numbered commands walk you through each step:
|
|
269
303
|
|
|
270
304
|
```
|
|
271
|
-
/fire-
|
|
272
|
-
/fire-2-plan 1
|
|
305
|
+
/fire-1a-new → Start your project (asks you questions, creates the plan)
|
|
306
|
+
/fire-2-plan 1 → Plan the first phase of work
|
|
273
307
|
/fire-3-execute 1 → Build it (Claude does the coding)
|
|
274
308
|
/fire-4-verify 1 → Check that everything actually works
|
|
275
309
|
/fire-5-handoff → Save your progress before closing
|
|
@@ -277,12 +311,17 @@ Claude will ask you a few simple questions about your project, then set everythi
|
|
|
277
311
|
```
|
|
278
312
|
|
|
279
313
|
**Want Claude to handle everything automatically?**
|
|
280
|
-
After `/fire-
|
|
314
|
+
After `/fire-1a-new`, just run:
|
|
281
315
|
```bash
|
|
282
316
|
/fire-autonomous
|
|
283
317
|
```
|
|
284
318
|
Claude will plan, build, and verify every phase without you having to type each command.
|
|
285
319
|
|
|
320
|
+
**Have a messy existing project?** Rebuild it clean:
|
|
321
|
+
```bash
|
|
322
|
+
/fire-phoenix --source ./my-app
|
|
323
|
+
```
|
|
324
|
+
|
|
286
325
|
---
|
|
287
326
|
|
|
288
327
|
## How Does It Compare?
|
|
@@ -293,32 +332,35 @@ Claude will plan, build, and verify every phase without you having to type each
|
|
|
293
332
|
|
|
294
333
|
| Feature | What It Does |
|
|
295
334
|
|---------|-------------|
|
|
296
|
-
|
|
|
297
|
-
|
|
|
335
|
+
| 46 slash commands | Every task has a dedicated command — no guessing |
|
|
336
|
+
| 478+ skills library | Proven patterns for auth, payments, APIs, and more |
|
|
337
|
+
| Phoenix Rebuild | Reverse-engineer messy code intent, rebuild clean from scratch |
|
|
298
338
|
| Breath-based parallelism | Independent tasks run at the same time |
|
|
299
|
-
|
|
|
339
|
+
| Tiered verification | Fast gate → must-haves → 70-point WARRIOR checklist |
|
|
300
340
|
| Session handoffs | Claude remembers everything between sessions |
|
|
301
|
-
| Circuit breaker |
|
|
341
|
+
| Circuit breaker intelligence | 6-type stuck classification with kill conditions |
|
|
302
342
|
| Auto skill extraction | Useful patterns discovered during work get saved automatically |
|
|
303
343
|
| Playwright E2E testing | Automated browser testing built in |
|
|
304
344
|
| Learncoding mode | Walk through any codebase step-by-step to learn it |
|
|
305
345
|
| Security scanning | Detect prompt injection, OWASP vulnerabilities, credential leaks |
|
|
346
|
+
| GoF design patterns | All 22 patterns mapped to AI agent architecture |
|
|
306
347
|
|
|
307
348
|
---
|
|
308
349
|
|
|
309
|
-
## All
|
|
350
|
+
## All 46 Commands
|
|
310
351
|
|
|
311
|
-
Commands are grouped into
|
|
352
|
+
Commands are grouped into 8 tiers. You only need Tier 1 for most projects.
|
|
312
353
|
|
|
313
354
|
| Tier | Purpose | Key Commands |
|
|
314
355
|
|------|---------|-------------|
|
|
315
|
-
| 1 — Core Workflow | The main pipeline | `/fire-
|
|
316
|
-
| 2 — Autonomous | Full autopilot | `/fire-autonomous`, `/fire-loop` |
|
|
356
|
+
| 1 — Core Workflow | The main pipeline | `/fire-1a-new` through `/fire-6-resume` |
|
|
357
|
+
| 2 — Autonomous | Full autopilot + Phoenix Rebuild | `/fire-autonomous`, `/fire-phoenix`, `/fire-loop` |
|
|
317
358
|
| 3 — Debug & Discover | Investigate problems | `/fire-debug`, `/fire-map-codebase` |
|
|
318
359
|
| 4 — Verification | Quality gates | `/fire-7-review`, `/fire-verify-uat` |
|
|
319
360
|
| 5 — Skills | Manage the pattern library | `/fire-search`, `/fire-add-new-skill` |
|
|
320
361
|
| 6 — Analytics & PM | Track progress | `/fire-dashboard`, `/fire-todos` |
|
|
321
362
|
| 7 — Milestones | Long-term projects | `/fire-new-milestone`, `/fire-complete-milestone` |
|
|
363
|
+
| 8 — Advanced | Specialized tools | `/fire-cost`, `/fire-execute-plan` |
|
|
322
364
|
|
|
323
365
|
See [COMMAND-REFERENCE.md](./COMMAND-REFERENCE.md) for the complete list with descriptions.
|
|
324
366
|
|
|
@@ -348,7 +390,7 @@ You can also create your own skills as you work. When Claude discovers a useful
|
|
|
348
390
|
| File | What It Covers |
|
|
349
391
|
|------|---------------|
|
|
350
392
|
| [QUICK-START.md](./QUICK-START.md) | Step-by-step walkthrough of your first project |
|
|
351
|
-
| [COMMAND-REFERENCE.md](./COMMAND-REFERENCE.md) | All
|
|
393
|
+
| [COMMAND-REFERENCE.md](./COMMAND-REFERENCE.md) | All 46 commands with descriptions |
|
|
352
394
|
| [references/warrior-principles.md](./references/warrior-principles.md) | The WARRIOR operating principles — what they are and why they matter |
|
|
353
395
|
| [DOMINION-FLOW-OVERVIEW.md](./DOMINION-FLOW-OVERVIEW.md) | Full system diagrams and architecture |
|
|
354
396
|
| [ARCHITECTURE-DIAGRAM.md](./ARCHITECTURE-DIAGRAM.md) | Visual overview of how everything connects |
|
|
@@ -385,6 +427,21 @@ View the full interactive version: **[Dominion Flow Architecture (HTML)](./docs/
|
|
|
385
427
|
|
|
386
428
|
---
|
|
387
429
|
|
|
430
|
+
## Privacy Policy
|
|
431
|
+
|
|
432
|
+
Dominion Flow runs entirely on your local machine. It does not:
|
|
433
|
+
|
|
434
|
+
- Collect, transmit, or store any personal data
|
|
435
|
+
- Send analytics or telemetry of any kind
|
|
436
|
+
- Connect to external servers (unless you optionally configure Qdrant or Ollama locally)
|
|
437
|
+
- Access any files outside your project directory and `~/.claude/plugins/`
|
|
438
|
+
|
|
439
|
+
All session memory, handoff files, and skills remain on your computer. The optional Qdrant vector database runs locally on your machine via Docker — no cloud services involved.
|
|
440
|
+
|
|
441
|
+
**In short: your code and data never leave your machine.**
|
|
442
|
+
|
|
443
|
+
---
|
|
444
|
+
|
|
388
445
|
## License
|
|
389
446
|
|
|
390
447
|
MIT License — Copyright (c) 2026 ThierryN
|