lytos-cli 0.2.0 → 0.2.1
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/README.md +84 -19
- package/package.json +2 -1
package/README.md
CHANGED
|
@@ -1,36 +1,53 @@
|
|
|
1
1
|
# Lytos — CLI
|
|
2
2
|
|
|
3
3
|
[](https://github.com/getlytos/lytos-cli/actions/workflows/ci.yml)
|
|
4
|
-
[](https://www.npmjs.com/package/lytos-cli)
|
|
4
|
+
[](https://www.npmjs.com/package/lytos-cli)
|
|
5
5
|
|
|
6
6
|
> The command-line tool for [Lytos](https://github.com/getlytos/lytos-method) — a human-first method for working with AI agents.
|
|
7
7
|
|
|
8
|
+
**Lytos** gives your AI agents a structured context: what the project is, how to work, what "done" means, what's in progress, and what's been learned. Everything in markdown. No vendor lock-in, no API, no account.
|
|
9
|
+
|
|
8
10
|
---
|
|
9
11
|
|
|
10
|
-
##
|
|
12
|
+
## The problem
|
|
11
13
|
|
|
12
|
-
|
|
13
|
-
|
|
14
|
-
|
|
14
|
+
AI agents (Claude, Cursor, GPT...) are powerful but stateless. Every session starts from zero. They don't know your project's conventions, your sprint priorities, or what was tried last week.
|
|
15
|
+
|
|
16
|
+
**Lytos** solves this by creating a `.lytos/` directory in your project — a structured context that any AI can read, and that you own.
|
|
17
|
+
|
|
18
|
+
---
|
|
15
19
|
|
|
16
|
-
|
|
20
|
+
## The 5 pillars
|
|
21
|
+
|
|
22
|
+
`lytos init` scaffolds a `.lytos/` directory built on 5 pillars:
|
|
23
|
+
|
|
24
|
+
| Pillar | Purpose | File / Directory |
|
|
25
|
+
|--------|---------|-----------------|
|
|
26
|
+
| **Intent** | Why the project exists — its constitution | `manifest.md` |
|
|
27
|
+
| **Design** | Procedures for recurring tasks (code review, testing, deployment...) | `skills/` |
|
|
28
|
+
| **Standards** | Non-negotiable quality criteria | `rules/` |
|
|
29
|
+
| **Progress** | Issues, sprint, what's moving and what's blocked | `issue-board/` |
|
|
30
|
+
| **Memory** | What's been learned — sovereign, portable, model-independent | `memory/` |
|
|
31
|
+
|
|
32
|
+
These 5 pillars are the method. The AI agent reads them at the start of each session and follows them.
|
|
33
|
+
|
|
34
|
+
---
|
|
35
|
+
|
|
36
|
+
## Install
|
|
17
37
|
|
|
18
38
|
```bash
|
|
19
|
-
|
|
20
|
-
lytos board # Regenerate BOARD.md from issue frontmatter
|
|
39
|
+
npm install -g lytos-cli
|
|
21
40
|
```
|
|
22
41
|
|
|
23
42
|
Or use without installing:
|
|
24
43
|
|
|
25
44
|
```bash
|
|
26
|
-
npx lytos init
|
|
45
|
+
npx lytos-cli init
|
|
27
46
|
```
|
|
28
47
|
|
|
29
48
|
---
|
|
30
49
|
|
|
31
|
-
##
|
|
32
|
-
|
|
33
|
-
One command to install the method, one command to validate your setup, one command to see your sprint.
|
|
50
|
+
## Commands
|
|
34
51
|
|
|
35
52
|
| Command | What it does |
|
|
36
53
|
|---------|-------------|
|
|
@@ -42,22 +59,70 @@ One command to install the method, one command to validate your setup, one comma
|
|
|
42
59
|
|
|
43
60
|
---
|
|
44
61
|
|
|
62
|
+
## What `lytos init` generates
|
|
63
|
+
|
|
64
|
+
```
|
|
65
|
+
your-project/
|
|
66
|
+
└── .lytos/
|
|
67
|
+
├── manifest.md # Intent — project identity and constraints
|
|
68
|
+
├── LYTOS.md # Method reference
|
|
69
|
+
├── sprint.md # Current sprint
|
|
70
|
+
├── skills/ # Design — 9 reusable procedures
|
|
71
|
+
│ ├── session-start.md
|
|
72
|
+
│ ├── code-structure.md
|
|
73
|
+
│ ├── code-review.md
|
|
74
|
+
│ ├── testing.md
|
|
75
|
+
│ ├── documentation.md
|
|
76
|
+
│ ├── git-workflow.md
|
|
77
|
+
│ ├── deployment.md
|
|
78
|
+
│ ├── security.md
|
|
79
|
+
│ └── api-design.md
|
|
80
|
+
├── rules/ # Standards — quality criteria
|
|
81
|
+
│ └── default-rules.md
|
|
82
|
+
├── issue-board/ # Progress — kanban board
|
|
83
|
+
│ ├── BOARD.md
|
|
84
|
+
│ ├── 0-icebox/
|
|
85
|
+
│ ├── 1-backlog/
|
|
86
|
+
│ ├── 2-sprint/
|
|
87
|
+
│ ├── 3-in-progress/
|
|
88
|
+
│ ├── 4-review/
|
|
89
|
+
│ └── 5-done/
|
|
90
|
+
└── memory/ # Memory — accumulated knowledge
|
|
91
|
+
├── MEMORY.md
|
|
92
|
+
└── cortex/
|
|
93
|
+
├── architecture.md
|
|
94
|
+
├── patterns.md
|
|
95
|
+
├── bugs.md
|
|
96
|
+
└── ...
|
|
97
|
+
```
|
|
98
|
+
|
|
99
|
+
---
|
|
100
|
+
|
|
101
|
+
## Design principles
|
|
102
|
+
|
|
103
|
+
- **Offline-first** — no network needed (except `lytos init` to download templates)
|
|
104
|
+
- **Zero lock-in** — plain markdown files, works with any AI tool
|
|
105
|
+
- **No telemetry** — no tracking, no analytics, ever
|
|
106
|
+
- **Human-first** — the human defines the method, the AI follows it
|
|
107
|
+
|
|
108
|
+
---
|
|
109
|
+
|
|
45
110
|
## Built with Lytos
|
|
46
111
|
|
|
47
112
|
This project uses Lytos to develop itself. The `.lytos/` directory contains the real manifest, sprint, issues, and memory for this project — not templates.
|
|
48
113
|
|
|
49
|
-
If you want to contribute, open this repo in Claude Code and say: **"Help me understand this project."**
|
|
50
|
-
|
|
51
114
|
---
|
|
52
115
|
|
|
53
|
-
##
|
|
116
|
+
## Links
|
|
117
|
+
|
|
118
|
+
- [Lytos Method](https://github.com/getlytos/lytos-method) — the method itself
|
|
119
|
+
- [Documentation](https://github.com/getlytos/lytos-website) — full docs (EN/FR)
|
|
54
120
|
|
|
55
|
-
|
|
121
|
+
---
|
|
56
122
|
|
|
57
|
-
|
|
58
|
-
- X: [@fred](https://x.com/fred)
|
|
123
|
+
## Author
|
|
59
124
|
|
|
60
|
-
|
|
125
|
+
Created by **Frederic Galline** — [ubeez.com](https://ubeez.com)
|
|
61
126
|
|
|
62
127
|
---
|
|
63
128
|
|
package/package.json
CHANGED
|
@@ -1,9 +1,10 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "lytos-cli",
|
|
3
|
-
"version": "0.2.
|
|
3
|
+
"version": "0.2.1",
|
|
4
4
|
"description": "CLI tool for Lytos — a human-first method for working with AI agents",
|
|
5
5
|
"type": "module",
|
|
6
6
|
"bin": {
|
|
7
|
+
"lyt": "dist/cli.js",
|
|
7
8
|
"lytos": "dist/cli.js",
|
|
8
9
|
"lytos-cli": "dist/cli.js"
|
|
9
10
|
},
|