luciazero 2.4.2 → 2.4.3

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 (2) hide show
  1. package/README.md +72 -11
  2. package/package.json +9 -2
package/README.md CHANGED
@@ -2,29 +2,68 @@
2
2
  <img src="https://cdn.jsdelivr.net/gh/ohm41321/luciazero@37cb470e2b7c704ff32f3a46dbb125e312875960/docs/assets/lucia.png" width="220" alt="Lucia — Luciazero's mascot">
3
3
  <h1>Luciazero</h1>
4
4
  <p>
5
- <strong>Evidence-first discipline for coding agents.</strong><br>
5
+ <strong>Make coding agents prove their work.</strong><br>
6
6
  <code>plan → change → verify → fix</code>
7
7
  </p>
8
8
  <p>
9
9
  <a href="https://www.npmjs.com/package/luciazero"><img src="https://img.shields.io/npm/v/luciazero" alt="npm version"></a>
10
+ <a href="https://www.npmjs.com/package/luciazero"><img src="https://img.shields.io/npm/dw/luciazero" alt="npm weekly downloads"></a>
10
11
  <a href="https://github.com/ohm41321/luciazero/actions/workflows/ci.yml"><img src="https://github.com/ohm41321/luciazero/actions/workflows/ci.yml/badge.svg" alt="CI status"></a>
12
+ <a href="https://github.com/ohm41321/luciazero/stargazers"><img src="https://img.shields.io/github/stars/ohm41321/luciazero?style=flat" alt="GitHub stars"></a>
13
+ <a href="https://github.com/ohm41321/luciazero/network/members"><img src="https://img.shields.io/github/forks/ohm41321/luciazero?style=flat" alt="GitHub forks"></a>
11
14
  <a href="https://github.com/ohm41321/luciazero/blob/main/LICENSE"><img src="https://img.shields.io/github/license/ohm41321/luciazero" alt="MIT license"></a>
12
15
  </p>
13
16
  </div>
14
17
 
15
18
  **English** · [ภาษาไทย](https://github.com/ohm41321/luciazero/blob/main/README.th.md)
16
19
 
17
- Luciazero is a lightweight discipline layer for coding agents. It works with
18
- Claude Code, Codex CLI, and compatible skill runtimes through plugin, CLI, or
19
- skills-only installs.
20
+ <p align="center">
21
+ <strong>11 skills</strong> · <strong>Relay fixture 6/6</strong> · <strong>Claude + Codex</strong> · <strong>MIT</strong>
22
+ </p>
23
+ <p align="center"><sub>Relay 6/6 is a mechanical protocol check; behavioral results are reported separately.</sub></p>
24
+
25
+ Luciazero is the verification and handoff layer for Claude Code, Codex CLI, and
26
+ compatible skill runtimes. It helps agents prove tests, preserve scope, and
27
+ move unfinished work with evidence.
20
28
 
21
29
  > Done is proven by a command, not by my judgment. If no verification command
22
30
  > exists, that is the first bug.
23
31
 
24
- It ships a compact [9-rule doctrine](https://github.com/ohm41321/luciazero/blob/main/claude/luciazero.md), 11 on-demand skills,
25
- verification hooks, a risk-routed reviewer, and an eval harness. It is designed
26
- to make completion claims auditable—not to replace an agent runtime or become an
27
- overnight orchestrator.
32
+ Luciazero is for teams that want coding agents to prove work instead of merely
33
+ report it. It helps prevent false-green verification and dropped scope,
34
+ preserves lessons from repeated dead ends, and carries context across handoffs
35
+ when `/lucia-relay` is used. It is a discipline layer, not an agent runtime.
36
+
37
+ ## Start in 30 seconds
38
+
39
+ Choose the path that matches your agent, then start a new session:
40
+
41
+ **Claude Code · full pack**
42
+
43
+ ```text
44
+ /plugin marketplace add ohm41321/luciazero
45
+ /plugin install luciazero@luciazero
46
+ ```
47
+
48
+ Run `/luciazero:ready` in the new session.
49
+
50
+ **Codex CLI · doctrine + skills**
51
+
52
+ ```bash
53
+ npx luciazero codex
54
+ ```
55
+
56
+ Run `$ready` in the new session.
57
+
58
+ **Skills-only · any compatible agent**
59
+
60
+ ```bash
61
+ npx skills add ohm41321/luciazero
62
+ ```
63
+
64
+ Invoke the installed `ready` skill using your agent's syntax (for Codex CLI,
65
+ use `$ready`). This path intentionally does not install the doctrine, reviewer,
66
+ or hooks.
28
67
 
29
68
  ## See the loop
30
69
 
@@ -40,6 +79,20 @@ This GIF is driven by the shipped hooks, not a mockup:
40
79
  ✅ verify 3m → the latest check passed three minutes ago
41
80
  ```
42
81
 
82
+ Run the same shipped hook driver locally—no model or API is required:
83
+
84
+ ```bash
85
+ bash docs/assets/statusline-demo.sh
86
+ ```
87
+
88
+ For the cross-session handoff demo:
89
+
90
+ ```bash
91
+ bash docs/assets/relay-demo.sh
92
+ ```
93
+
94
+ Both scripts use throwaway directories and exercise the real implementation.
95
+
43
96
  ## What it protects
44
97
 
45
98
  | Failure mode | Mechanism |
@@ -151,11 +204,13 @@ Other install channels use their own updater:
151
204
 
152
205
  ```bash
153
206
  claude plugin update luciazero@luciazero # then run /reload-plugins
154
- npx skills update # review the scope prompt
207
+ npx skills update # every installed skill in the selected scope
208
+ npx skills update done -g # only the global "done" skill
155
209
  ```
156
210
 
157
211
  The skills command updates every installed skill in the selected scope, not
158
- only Luciazero; review its prompt before confirming.
212
+ only Luciazero; review its prompt before confirming. Use the targeted form when
213
+ you only want to refresh one skill.
159
214
 
160
215
  Claude Code can auto-update the plugin at startup: open `/plugin` →
161
216
  **Marketplaces** → **luciazero** → **Enable auto-update**. Third-party
@@ -164,7 +219,8 @@ GitHub **Watch → Custom → Releases**.
164
219
 
165
220
  ## Skills at a glance
166
221
 
167
- Run `/ready` first; the rest activate when their moment arrives.
222
+ Invoke `ready` first (`/ready` for slash-based agents, `$ready` in Codex); the
223
+ rest activate when their moment arrives.
168
224
 
169
225
  | Moment | Skill | Result |
170
226
  |---|---|---|
@@ -191,6 +247,10 @@ separate passes.
191
247
 
192
248
  ## Evidence & limitations
193
249
 
250
+ The measurements below are early and model/task-specific. The linked raw rows
251
+ and methodology are the source of truth; they are not a promise of uplift on
252
+ every repository or model.
253
+
194
254
  <!-- BEGIN GENERATED: benchmark-evidence -->
195
255
 
196
256
  ### Claude results
@@ -273,6 +333,7 @@ More detail:
273
333
  - [Benchmark results and GPT plan](https://github.com/ohm41321/luciazero/blob/main/docs/benchmark.md)
274
334
  - [Raw campaign registry](https://github.com/ohm41321/luciazero/blob/main/eval/results/campaigns.json)
275
335
  - [Experiment log](https://github.com/ohm41321/luciazero/blob/main/docs/experiments.md)
336
+ - [Launch kit](https://github.com/ohm41321/luciazero/blob/main/docs/launch-kit.md)
276
337
  - [Contributing](https://github.com/ohm41321/luciazero/blob/main/CONTRIBUTING.md)
277
338
  - [Publishing](https://github.com/ohm41321/luciazero/blob/main/docs/publishing.md)
278
339
  - [Changelog](https://github.com/ohm41321/luciazero/blob/main/CHANGELOG.md)
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "luciazero",
3
- "version": "2.4.2",
3
+ "version": "2.4.3",
4
4
  "description": "Verification-first discipline for coding agents (Claude Code + Codex CLI): 9-rule doctrine, 11 skills, risk-routed reviewer, fail-open enforcement hooks. npx luciazero installs it.",
5
5
  "repository": {
6
6
  "type": "git",
@@ -32,10 +32,17 @@
32
32
  "keywords": [
33
33
  "claude-code",
34
34
  "codex",
35
+ "ai-agent",
36
+ "ai-agents",
35
37
  "agent-skills",
38
+ "coding-agent",
39
+ "claude-plugin",
40
+ "context-handoff",
36
41
  "verification",
37
42
  "doctrine",
38
43
  "hooks",
39
- "agentic-engineering"
44
+ "agentic-engineering",
45
+ "agent-workflow",
46
+ "test-discipline"
40
47
  ]
41
48
  }