commit-analyzer 1.1.4 → 1.1.5
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 +164 -82
- package/package.json +1 -1
- package/prompt.md +2 -2
- package/src/3.presentation/cli-application.ts +1 -1
package/README.md
CHANGED
|
@@ -1,43 +1,29 @@
|
|
|
1
1
|
# Git Commit Analyzer
|
|
2
2
|
|
|
3
|
-
A TypeScript/Node.js program that analyzes git commits and generates categorized
|
|
3
|
+
A TypeScript/Node.js program that analyzes git commits and generates categorized
|
|
4
|
+
summaries using Claude CLI.
|
|
4
5
|
|
|
5
6
|
## Features
|
|
6
7
|
|
|
7
8
|
- Extract commit details (message, date, diff) from git repositories
|
|
8
|
-
- Categorize commits using LLM analysis into:
|
|
9
|
-
|
|
10
|
-
- Generate
|
|
11
|
-
-
|
|
9
|
+
- Categorize commits using LLM analysis into:
|
|
10
|
+
`tweak`, `feature`, or `process`
|
|
11
|
+
- Generate CSV reports with timestamp, category, summary, and description
|
|
12
|
+
- Generate condensed markdown reports from CSV data for stakeholder
|
|
13
|
+
communication
|
|
14
|
+
- Support for multiple LLM models (Claude, Gemini, OpenAI) with automatic
|
|
15
|
+
detection
|
|
12
16
|
- Support for batch processing multiple commits
|
|
13
17
|
- Automatically filters out merge commits for cleaner analysis
|
|
14
18
|
- Robust error handling and validation
|
|
15
19
|
|
|
16
|
-
## Prerequisites
|
|
17
|
-
|
|
18
|
-
This tool requires Bun runtime. Install it globally:
|
|
19
|
-
|
|
20
|
-
```bash
|
|
21
|
-
# Install bun globally
|
|
22
|
-
curl -fsSL https://bun.sh/install | bash
|
|
23
|
-
# or
|
|
24
|
-
npm install -g bun
|
|
25
|
-
```
|
|
26
|
-
|
|
27
|
-
## Installation
|
|
28
|
-
|
|
29
|
-
```bash
|
|
30
|
-
npm install
|
|
31
|
-
bun link
|
|
32
|
-
```
|
|
33
|
-
|
|
34
|
-
After linking, you can use `commit-analyzer` command globally.
|
|
35
20
|
|
|
36
21
|
## Usage
|
|
37
22
|
|
|
38
23
|
### Default Behavior
|
|
39
24
|
|
|
40
|
-
When run without arguments, the program analyzes all commits authored by the
|
|
25
|
+
When run without arguments, the program analyzes all commits authored by the
|
|
26
|
+
current user:
|
|
41
27
|
|
|
42
28
|
```bash
|
|
43
29
|
# Analyze all your commits in the current repository
|
|
@@ -56,9 +42,6 @@ npx commit-analyzer --author user@example.com
|
|
|
56
42
|
# Analyze specific commits
|
|
57
43
|
npx commit-analyzer abc123 def456 ghi789
|
|
58
44
|
|
|
59
|
-
# Read commits from file
|
|
60
|
-
npx commit-analyzer --file commits.txt
|
|
61
|
-
|
|
62
45
|
# Specify output file with default behavior
|
|
63
46
|
npx commit-analyzer --output analysis.csv --limit 20
|
|
64
47
|
|
|
@@ -69,32 +52,46 @@ npx commit-analyzer --report --input-csv analysis.csv
|
|
|
69
52
|
npx commit-analyzer --report --limit 50
|
|
70
53
|
|
|
71
54
|
# Use specific LLM model
|
|
72
|
-
npx commit-analyzer --
|
|
55
|
+
npx commit-analyzer --llm claude --limit 10
|
|
73
56
|
```
|
|
74
57
|
|
|
75
58
|
### Options
|
|
76
59
|
|
|
77
|
-
- `-o, --output <file>`:
|
|
78
|
-
|
|
79
|
-
|
|
80
|
-
-
|
|
81
|
-
|
|
82
|
-
|
|
83
|
-
- `-
|
|
84
|
-
|
|
85
|
-
-
|
|
86
|
-
|
|
87
|
-
-
|
|
88
|
-
|
|
89
|
-
|
|
90
|
-
|
|
91
|
-
|
|
92
|
-
|
|
93
|
-
|
|
94
|
-
|
|
95
|
-
|
|
96
|
-
|
|
97
|
-
|
|
60
|
+
- `-o, --output <file>`:
|
|
61
|
+
Output file (default:
|
|
62
|
+
`results/commits.csv` for analysis, `results/report.md` for reports)
|
|
63
|
+
- `--output-dir <dir>`:
|
|
64
|
+
Output directory for CSV and report files (default:
|
|
65
|
+
current directory)
|
|
66
|
+
- `-a, --author <email>`:
|
|
67
|
+
Filter commits by author email (defaults to current user)
|
|
68
|
+
- `-l, --limit <number>`:
|
|
69
|
+
Limit number of commits to analyze
|
|
70
|
+
- `--llm <model>`:
|
|
71
|
+
LLM model to use (claude, gemini, openai)
|
|
72
|
+
- `-r, --resume`:
|
|
73
|
+
Resume from last checkpoint if available
|
|
74
|
+
- `-c, --clear`:
|
|
75
|
+
Clear any existing progress checkpoint
|
|
76
|
+
- `--report`:
|
|
77
|
+
Generate condensed markdown report from existing CSV
|
|
78
|
+
- `--input-csv <file>`:
|
|
79
|
+
Input CSV file to read for report generation
|
|
80
|
+
- `-v, --verbose`:
|
|
81
|
+
Enable verbose logging (shows detailed error information)
|
|
82
|
+
- `--since <date>`:
|
|
83
|
+
Only analyze commits since this date (YYYY-MM-DD, '1 week ago', '2024-01-01')
|
|
84
|
+
- `--until <date>`:
|
|
85
|
+
Only analyze commits until this date (YYYY-MM-DD, '1 day ago', '2024-12-31')
|
|
86
|
+
- `--no-cache`:
|
|
87
|
+
Disable caching of analysis results
|
|
88
|
+
- `--batch-size <number>`:
|
|
89
|
+
Number of commits to process per batch (default:
|
|
90
|
+
1 for sequential processing)
|
|
91
|
+
- `-h, --help`:
|
|
92
|
+
Display help
|
|
93
|
+
- `-V, --version`:
|
|
94
|
+
Display version
|
|
98
95
|
|
|
99
96
|
## Output Formats
|
|
100
97
|
|
|
@@ -102,36 +99,46 @@ ghi789jkl012
|
|
|
102
99
|
|
|
103
100
|
The program generates a CSV file with the following columns:
|
|
104
101
|
|
|
105
|
-
- `
|
|
106
|
-
|
|
107
|
-
- `
|
|
108
|
-
|
|
102
|
+
- `timestamp`:
|
|
103
|
+
ISO 8601 timestamp of the commit (e.g., `2025-08-28T11:14:40.000Z`)
|
|
104
|
+
- `category`:
|
|
105
|
+
One of `tweak`, `feature`, or `process`
|
|
106
|
+
- `summary`:
|
|
107
|
+
One-line description (max 80 characters)
|
|
108
|
+
- `description`:
|
|
109
|
+
Detailed explanation (2-3 sentences)
|
|
109
110
|
|
|
110
111
|
### Markdown Report Output
|
|
111
112
|
|
|
112
|
-
When using the `--report` option, the program generates a condensed markdown
|
|
113
|
+
When using the `--report` option, the program generates a condensed markdown
|
|
114
|
+
report that:
|
|
113
115
|
|
|
114
116
|
- Groups commits by year (most recent first)
|
|
115
|
-
- Organizes by categories:
|
|
117
|
+
- Organizes by categories:
|
|
118
|
+
Features, Processes, Tweaks & Bug Fixes
|
|
116
119
|
- Consolidates similar items for stakeholder readability
|
|
117
120
|
- Includes commit count statistics
|
|
118
|
-
- Uses professional language suitable for both technical and non-technical
|
|
121
|
+
- Uses professional language suitable for both technical and non-technical
|
|
122
|
+
audiences
|
|
119
123
|
|
|
120
124
|
## Requirements
|
|
121
125
|
|
|
122
|
-
- Node.js 18+ with TypeScript support
|
|
126
|
+
- Node.js 18+ with TypeScript support (Bun runtime recommended)
|
|
123
127
|
- Git repository (must be run within a git repository)
|
|
124
128
|
- At least one supported LLM CLI tool:
|
|
125
129
|
- Claude CLI (`claude`) - recommended, defaults to Sonnet model
|
|
126
130
|
- Gemini CLI (`gemini`)
|
|
127
|
-
-
|
|
131
|
+
- OpenAI CLI (`codex`)
|
|
128
132
|
- Valid git commit hashes (when specifying commits manually)
|
|
129
133
|
|
|
130
134
|
## Categories
|
|
131
135
|
|
|
132
|
-
- **tweak**:
|
|
133
|
-
|
|
134
|
-
- **
|
|
136
|
+
- **tweak**:
|
|
137
|
+
Minor adjustments, bug fixes, small improvements
|
|
138
|
+
- **feature**:
|
|
139
|
+
New functionality, major additions
|
|
140
|
+
- **process**:
|
|
141
|
+
Build system, CI/CD, tooling, configuration changes
|
|
135
142
|
|
|
136
143
|
## Error Handling
|
|
137
144
|
|
|
@@ -151,7 +158,8 @@ The tool automatically:
|
|
|
151
158
|
- **Stops processing after a commit fails all retry attempts**
|
|
152
159
|
- Exports partial results to the CSV file before exiting
|
|
153
160
|
|
|
154
|
-
If the process stops (e.g., after 139 commits due to API failure), you can
|
|
161
|
+
If the process stops (e.g., after 139 commits due to API failure), you can
|
|
162
|
+
resume from where it left off:
|
|
155
163
|
|
|
156
164
|
```bash
|
|
157
165
|
# Resume from last checkpoint
|
|
@@ -165,13 +173,12 @@ npx commit-analyzer --resume
|
|
|
165
173
|
```
|
|
166
174
|
|
|
167
175
|
The checkpoint file (`.commit-analyzer/progress.json`) contains:
|
|
176
|
+
|
|
168
177
|
- List of all commits to process
|
|
169
178
|
- Successfully processed commits (including failed ones to skip on resume)
|
|
170
179
|
- Analyzed commit data (only successful ones)
|
|
171
180
|
- Output file location
|
|
172
181
|
|
|
173
|
-
**Important**: When a commit fails after all retries (default 3), the process stops immediately to prevent wasting API calls. The successfully analyzed commits up to that point are saved to the CSV file.
|
|
174
|
-
|
|
175
182
|
### Application Data Directory
|
|
176
183
|
|
|
177
184
|
The tool creates a `.commit-analyzer/` directory to store internal files:
|
|
@@ -185,23 +192,80 @@ The tool creates a `.commit-analyzer/` directory to store internal files:
|
|
|
185
192
|
└── ...
|
|
186
193
|
```
|
|
187
194
|
|
|
188
|
-
- **Progress checkpoint**:
|
|
189
|
-
|
|
195
|
+
- **Progress checkpoint**:
|
|
196
|
+
Enables resuming interrupted analysis sessions
|
|
197
|
+
- **Analysis cache**:
|
|
198
|
+
Stores LLM analysis results to avoid re-processing the same commits (TTL:
|
|
199
|
+
30 days)
|
|
190
200
|
|
|
191
201
|
Use `--no-cache` to disable caching if needed.
|
|
202
|
+
Use `--clear` to clear the cache and progress checkpoint.
|
|
203
|
+
|
|
204
|
+
### Date Filtering
|
|
205
|
+
|
|
206
|
+
The tool supports flexible date filtering using natural language or specific
|
|
207
|
+
dates:
|
|
208
|
+
|
|
209
|
+
```bash
|
|
210
|
+
# Analyze commits from the last week
|
|
211
|
+
npx commit-analyzer --since "1 week ago"
|
|
212
|
+
|
|
213
|
+
# Analyze commits from a specific date range
|
|
214
|
+
npx commit-analyzer --since "2024-01-01" --until "2024-12-31"
|
|
215
|
+
|
|
216
|
+
# Analyze commits from the beginning of the year
|
|
217
|
+
npx commit-analyzer --since "2024-01-01"
|
|
218
|
+
|
|
219
|
+
# Analyze commits up to a specific date
|
|
220
|
+
npx commit-analyzer --until "2024-06-30"
|
|
221
|
+
```
|
|
222
|
+
|
|
223
|
+
Date formats supported:
|
|
224
|
+
- Relative dates:
|
|
225
|
+
`"1 week ago"`, `"2 months ago"`, `"3 days ago"`
|
|
226
|
+
- ISO dates:
|
|
227
|
+
`"2024-01-01"`, `"2024-12-31"`
|
|
228
|
+
- Git-style dates:
|
|
229
|
+
Any format accepted by `git log --since` and `git log --until`
|
|
230
|
+
|
|
231
|
+
### Batch Processing
|
|
232
|
+
|
|
233
|
+
Control processing speed and resource usage with batch size options:
|
|
234
|
+
|
|
235
|
+
```bash
|
|
236
|
+
# Process commits one at a time (default, safest for rate limits)
|
|
237
|
+
npx commit-analyzer --batch-size 1
|
|
238
|
+
|
|
239
|
+
# Process multiple commits in parallel (faster but may hit rate limits)
|
|
240
|
+
npx commit-analyzer --batch-size 5 --limit 100
|
|
241
|
+
|
|
242
|
+
# Sequential processing for large datasets
|
|
243
|
+
npx commit-analyzer --batch-size 1 --limit 500
|
|
244
|
+
```
|
|
192
245
|
|
|
193
246
|
### Retry Logic
|
|
194
247
|
|
|
195
|
-
The tool includes automatic retry logic with exponential backoff for handling
|
|
248
|
+
The tool includes automatic retry logic with exponential backoff for handling
|
|
249
|
+
API failures when processing many commits.
|
|
250
|
+
This is especially useful when analyzing large numbers of commits that might
|
|
251
|
+
trigger rate limits.
|
|
196
252
|
|
|
197
253
|
#### Configuration
|
|
198
254
|
|
|
199
255
|
You can configure the retry behavior using environment variables:
|
|
200
256
|
|
|
201
|
-
- `LLM_MAX_RETRIES`:
|
|
202
|
-
|
|
203
|
-
|
|
204
|
-
- `
|
|
257
|
+
- `LLM_MAX_RETRIES`:
|
|
258
|
+
Maximum number of retry attempts (default:
|
|
259
|
+
3)
|
|
260
|
+
- `LLM_INITIAL_RETRY_DELAY`:
|
|
261
|
+
Initial delay between retries in milliseconds (default:
|
|
262
|
+
5000)
|
|
263
|
+
- `LLM_MAX_RETRY_DELAY`:
|
|
264
|
+
Maximum delay between retries in milliseconds (default:
|
|
265
|
+
30000)
|
|
266
|
+
- `LLM_RETRY_MULTIPLIER`:
|
|
267
|
+
Multiplier for exponential backoff (default:
|
|
268
|
+
2)
|
|
205
269
|
|
|
206
270
|
#### Examples
|
|
207
271
|
|
|
@@ -226,19 +290,19 @@ The retry mechanism automatically:
|
|
|
226
290
|
|
|
227
291
|
```bash
|
|
228
292
|
# Install dependencies
|
|
229
|
-
|
|
293
|
+
bun install
|
|
230
294
|
|
|
231
295
|
# Run in development mode
|
|
232
|
-
|
|
296
|
+
bun run dev
|
|
233
297
|
|
|
234
298
|
# Build for production
|
|
235
|
-
|
|
299
|
+
bun run build
|
|
236
300
|
|
|
237
301
|
# Run linting
|
|
238
|
-
|
|
302
|
+
bun run lint
|
|
239
303
|
|
|
240
304
|
# Type checking
|
|
241
|
-
|
|
305
|
+
bun run typecheck
|
|
242
306
|
```
|
|
243
307
|
|
|
244
308
|
## Examples
|
|
@@ -253,10 +317,6 @@ npx commit-analyzer --limit 20 --output my_analysis.csv
|
|
|
253
317
|
# Analyze commits by a specific team member
|
|
254
318
|
npx commit-analyzer --author teammate@company.com --limit 50
|
|
255
319
|
|
|
256
|
-
# Analyze specific commits
|
|
257
|
-
git log --oneline -5 | cut -d' ' -f1 > recent_commits.txt
|
|
258
|
-
npx commit-analyzer --file recent_commits.txt --output recent_analysis.csv
|
|
259
|
-
|
|
260
320
|
# Quick analysis of your recent work
|
|
261
321
|
npx commit-analyzer --limit 10
|
|
262
322
|
|
|
@@ -267,8 +327,30 @@ npx commit-analyzer --report --limit 100 --output yearly_analysis.csv
|
|
|
267
327
|
npx commit-analyzer --report --input-csv existing_analysis.csv --output team_report.md
|
|
268
328
|
|
|
269
329
|
# Use specific LLM model for analysis
|
|
270
|
-
npx commit-analyzer --
|
|
330
|
+
npx commit-analyzer --llm gemini --limit 25
|
|
271
331
|
|
|
272
332
|
# Resume interrupted analysis with progress tracking
|
|
273
333
|
npx commit-analyzer --resume
|
|
274
334
|
```
|
|
335
|
+
|
|
336
|
+
## Development
|
|
337
|
+
|
|
338
|
+
This tool requires the Bun runtime.
|
|
339
|
+
Install it globally:
|
|
340
|
+
|
|
341
|
+
```bash
|
|
342
|
+
# Install bun globally
|
|
343
|
+
curl -fsSL https://bun.sh/install | bash
|
|
344
|
+
# or
|
|
345
|
+
npm install -g bun
|
|
346
|
+
```
|
|
347
|
+
|
|
348
|
+
## Installation
|
|
349
|
+
|
|
350
|
+
```bash
|
|
351
|
+
bun install
|
|
352
|
+
bun build
|
|
353
|
+
bun link
|
|
354
|
+
```
|
|
355
|
+
|
|
356
|
+
After linking, you can use `commit-analyzer` command globally.
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "commit-analyzer",
|
|
3
|
-
"version": "1.1.
|
|
3
|
+
"version": "1.1.5",
|
|
4
4
|
"description": "Analyze git commits and generate categories, summaries, and descriptions for each commit. Optionally generate a yearly breakdown report of your commit history.",
|
|
5
5
|
"main": "dist/main.ts",
|
|
6
6
|
"bin": {
|
package/prompt.md
CHANGED
|
@@ -58,12 +58,12 @@
|
|
|
58
58
|
### Technical Implementation
|
|
59
59
|
|
|
60
60
|
- Use Node.js with TypeScript
|
|
61
|
-
- Extract
|
|
61
|
+
- Extract timestamp from git commit
|
|
62
62
|
|
|
63
63
|
Output Format:
|
|
64
64
|
|
|
65
65
|
CSV with headers:
|
|
66
|
-
|
|
66
|
+
timestamp,category,summary,description
|
|
67
67
|
|
|
68
68
|
The program should be robust, handle edge cases, and provide clear error
|
|
69
69
|
messages for invalid commits or API failures.
|
|
@@ -24,7 +24,7 @@ export interface CLIOptions {
|
|
|
24
24
|
}
|
|
25
25
|
|
|
26
26
|
export class CLIApplication {
|
|
27
|
-
private static readonly VERSION = "1.1.
|
|
27
|
+
private static readonly VERSION = "1.1.5"
|
|
28
28
|
private static readonly DEFAULT_COMMITS_OUTPUT_FILE = "results/commits.csv"
|
|
29
29
|
private static readonly DEFAULT_REPORT_OUTPUT_FILE = "results/report.md"
|
|
30
30
|
|