@codyswann/lisa 1.8.0 → 1.9.2
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.
|
@@ -109,24 +109,29 @@ For each task, use **TaskCreate** with:
|
|
|
109
109
|
[Or "N/A - no user-facing changes"]
|
|
110
110
|
|
|
111
111
|
## Verification
|
|
112
|
+
|
|
113
|
+
Every task MUST have empirical verification - a command that proves the work is done.
|
|
114
|
+
|
|
115
|
+
**Why?** Code that "looks correct" often isn't. The only way to know something works is to run it and observe the result. Verification isn't "I wrote the code" - it's running a command and seeing expected output.
|
|
116
|
+
|
|
112
117
|
**Type:** `test` | `ui-recording` | `test-coverage` | `api-test` | `manual-check` | `documentation`
|
|
113
118
|
|
|
114
119
|
| Type | When to Use | Example |
|
|
115
120
|
|------|-------------|---------|
|
|
116
|
-
| `test` | Run specific tests | `
|
|
117
|
-
| `ui-recording` | UI/UX changes | `
|
|
118
|
-
| `test-coverage` | Coverage threshold | `
|
|
119
|
-
| `api-test` | API endpoints |
|
|
120
|
-
| `documentation` | Docs, README | `cat path/to/doc.md` |
|
|
121
|
-
| `manual-check` | Config, setup |
|
|
121
|
+
| `test` | Run specific tests | `npm test -- src/services/user.spec.ts` |
|
|
122
|
+
| `ui-recording` | UI/UX changes | `npm run playwright:test ...` |
|
|
123
|
+
| `test-coverage` | Coverage threshold | `npm run test:cov -- --collectCoverageFrom='...'` |
|
|
124
|
+
| `api-test` | API endpoints | `curl -s http://localhost:3000/api/endpoint` |
|
|
125
|
+
| `documentation` | Docs, README | `cat path/to/doc.md \| grep "expected content"` |
|
|
126
|
+
| `manual-check` | Config, setup | `cat config.json \| jq '.setting'` |
|
|
122
127
|
|
|
123
128
|
**Proof Command:**
|
|
124
129
|
```bash
|
|
125
|
-
[Single command
|
|
130
|
+
[Single command that outputs observable proof of completion]
|
|
126
131
|
```
|
|
127
132
|
|
|
128
133
|
**Expected Output:**
|
|
129
|
-
[
|
|
134
|
+
[Exact or pattern-matched output that proves success]
|
|
130
135
|
|
|
131
136
|
## Learnings
|
|
132
137
|
During implementation, collect any discoveries valuable for future developers:
|
|
@@ -34,7 +34,7 @@ Based on input type:
|
|
|
34
34
|
|
|
35
35
|
2. Create `brief.md` in the project directory:
|
|
36
36
|
- Jira: populate with issue number, title, and description
|
|
37
|
-
- File:
|
|
37
|
+
- File: copy contents of the file to `brief.md`
|
|
38
38
|
- Text prompt: create `brief.md` with the prompt text as content
|
|
39
39
|
|
|
40
40
|
3. Create empty `findings.md` in the project directory
|
|
@@ -0,0 +1,37 @@
|
|
|
1
|
+
# Empirical Verification
|
|
2
|
+
|
|
3
|
+
Every task requires a **proof command** - a single command that empirically demonstrates the work is done.
|
|
4
|
+
|
|
5
|
+
## Core Principle
|
|
6
|
+
|
|
7
|
+
Never assume something works because the code "looks correct." Run a command, observe the output, compare to expected result.
|
|
8
|
+
|
|
9
|
+
## Verification Types
|
|
10
|
+
|
|
11
|
+
| Type | Use Case | Example |
|
|
12
|
+
|------|----------|---------|
|
|
13
|
+
| `test` | Unit/integration tests | `npm test -- path/to/file.spec.ts` |
|
|
14
|
+
| `api-test` | API endpoints | `curl -s localhost:3000/api/endpoint` |
|
|
15
|
+
| `test-coverage` | Coverage thresholds | `npm run test:cov -- --collectCoverageFrom=...` |
|
|
16
|
+
| `ui-recording` | UI changes | Playwright/Cypress test |
|
|
17
|
+
| `documentation` | Doc changes | `grep "expected" path/to/doc.md` |
|
|
18
|
+
| `manual-check` | Config/setup | `cat config.json \| jq '.setting'` |
|
|
19
|
+
|
|
20
|
+
## Task Completion Rules
|
|
21
|
+
|
|
22
|
+
1. **Run the proof command** before marking any task complete
|
|
23
|
+
2. **Compare output** to expected result
|
|
24
|
+
3. **If verification fails**: Fix and re-run, don't mark complete
|
|
25
|
+
4. **If verification blocked** (missing Docker, services, etc.): Mark as blocked, not complete
|
|
26
|
+
|
|
27
|
+
## Example
|
|
28
|
+
|
|
29
|
+
**Task**: Add health check endpoint
|
|
30
|
+
|
|
31
|
+
**Wrong verification**: "I added the route handler"
|
|
32
|
+
|
|
33
|
+
**Correct verification**:
|
|
34
|
+
```bash
|
|
35
|
+
curl -s http://localhost:3000/health | jq '.status'
|
|
36
|
+
```
|
|
37
|
+
**Expected**: `"ok"`
|
package/package.json
CHANGED
|
@@ -26,7 +26,7 @@
|
|
|
26
26
|
"knip": "knip",
|
|
27
27
|
"knip:fix": "knip --fix",
|
|
28
28
|
"sg:scan": "ast-grep scan",
|
|
29
|
-
"lisa:update": "npx @codyswann/lisa ."
|
|
29
|
+
"lisa:update": "npx @codyswann/lisa@latest ."
|
|
30
30
|
},
|
|
31
31
|
"devDependencies": {
|
|
32
32
|
"@ast-grep/cli": "^0.40.4",
|
|
@@ -80,7 +80,7 @@
|
|
|
80
80
|
"@ast-grep/cli"
|
|
81
81
|
],
|
|
82
82
|
"name": "@codyswann/lisa",
|
|
83
|
-
"version": "1.
|
|
83
|
+
"version": "1.9.2",
|
|
84
84
|
"description": "Claude Code governance framework that applies guardrails, guidance, and automated enforcement to projects",
|
|
85
85
|
"main": "dist/index.js",
|
|
86
86
|
"bin": {
|