aicommit2 2.0.8 → 2.1.0

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.
Files changed (3) hide show
  1. package/README.md +184 -158
  2. package/dist/cli.mjs +81 -71
  3. package/package.json +6 -6
package/README.md CHANGED
@@ -179,7 +179,7 @@ aicommit2 config set OPENAI.generate=3 GEMINI.temperature=0.5
179
179
 
180
180
  #### How to Configure in detail
181
181
 
182
- 1. Command-line arguments: **use the format** `--[ModelName].[SettingKey]=value`
182
+ 1. Command-line arguments: **use the format** `--[Model].[Key]=value`
183
183
  ```sh
184
184
  aicommit2 --OPENAI.locale="jp" --GEMINI.temperatue="0.5"
185
185
  ```
@@ -202,7 +202,7 @@ systemPromptPath="<your-prompt-path>"
202
202
  [GEMINI]
203
203
  key="<your-api-key>"
204
204
  generate=5
205
- ignoreBody=false
205
+ includeBody=true
206
206
 
207
207
  [OLLAMA]
208
208
  temperature=0.7
@@ -217,27 +217,29 @@ model[]=codestral
217
217
  The following settings can be applied to most models, but support may vary.
218
218
  Please check the documentation for each specific model to confirm which settings are supported.
219
219
 
220
- | Setting | Description | Default |
221
- |--------------------|---------------------------------------------------------------------|---------------|
222
- | `systemPrompt` | System Prompt text | - |
223
- | `systemPromptPath` | Path to system prompt file | - |
224
- | `exclude` | Files to exclude from AI analysis | - |
225
- | `type` | Type of commit message to generate | conventional |
226
- | `locale` | Locale for the generated commit messages | en |
227
- | `generate` | Number of commit messages to generate | 1 |
228
- | `logging` | Enable logging | true |
229
- | `ignoreBody` | Whether the commit message includes body | true |
230
- | `maxLength` | Maximum character length of the Subject of generated commit message | 50 |
231
- | `timeout` | Request timeout (milliseconds) | 10000 |
232
- | `temperature` | Model's creativity (0.0 - 2.0) | 0.7 |
233
- | `maxTokens` | Maximum number of tokens to generate | 1024 |
234
- | `topP` | Nucleus sampling | 1 |
220
+ | Setting | Description | Default |
221
+ |------------------------|---------------------------------------------------------------------|--------------|
222
+ | `systemPrompt` | System Prompt text | - |
223
+ | `systemPromptPath` | Path to system prompt file | - |
224
+ | `exclude` | Files to exclude from AI analysis | - |
225
+ | `type` | Type of commit message to generate | conventional |
226
+ | `locale` | Locale for the generated commit messages | en |
227
+ | `generate` | Number of commit messages to generate | 1 |
228
+ | `logging` | Enable logging | true |
229
+ | `includeBody` | Whether the commit message includes body | false |
230
+ | `maxLength` | Maximum character length of the Subject of generated commit message | 50 |
231
+ | `timeout` | Request timeout (milliseconds) | 10000 |
232
+ | `temperature` | Model's creativity (0.0 - 2.0) | 0.7 |
233
+ | `maxTokens` | Maximum number of tokens to generate | 1024 |
234
+ | `topP` | Nucleus sampling | 0.9 |
235
+ | `codeReview` | whether to include an automated code review in the process | false |
236
+ | `codeReviewPromptPath` | Path to code review prompt file | - |
235
237
 
236
238
  > 👉 **Tip:** To set the General Settings for each model, use the following command.
237
239
  > ```shell
238
240
  > aicommit2 config set OPENAI.locale="jp"
239
241
  > aicommit2 config set CODESTRAL.type="gitmoji"
240
- > aicommit2 config set GEMINI.ignoreBody=false
242
+ > aicommit2 config set GEMINI.includeBody=true
241
243
  > ```
242
244
 
243
245
  ##### systemPrompt
@@ -318,21 +320,21 @@ The log files will be stored in the `~/.aicommit2_log` directory(user's home).
318
320
  aicommit2 log removeAll
319
321
  ```
320
322
 
321
- ##### ignoreBody
323
+ ##### includeBody
322
324
 
323
- Default: `true`
325
+ Default: `false`
324
326
 
325
- This option determines whether the commit message includes body. If you want to include body in message, you can set it to `false`.
327
+ This option determines whether the commit message includes body. If you want to include body in message, you can set it to `true`.
326
328
 
327
329
  ```sh
328
- aicommit2 config set ignoreBody="false"
330
+ aicommit2 config set includeBody="true"
329
331
  ```
330
332
 
331
333
  ![ignore_body_false](https://github.com/tak-bro/aicommit2/blob/main/img/demo_body_min.gif?raw=true)
332
334
 
333
335
 
334
336
  ```sh
335
- aicommit2 config set ignoreBody="true"
337
+ aicommit2 config set includeBody="false"
336
338
  ```
337
339
 
338
340
  ![ignore_body_true](https://github.com/tak-bro/aicommit2/blob/main/img/ignore_body_true.png?raw=true)
@@ -379,7 +381,7 @@ aicommit2 config set maxTokens=3000
379
381
 
380
382
  ##### topP
381
383
 
382
- Default: `1`
384
+ Default: `0.9`
383
385
 
384
386
  Nucleus sampling, where the model considers the results of the tokens with top_p probability mass.
385
387
 
@@ -387,23 +389,53 @@ Nucleus sampling, where the model considers the results of the tokens with top_p
387
389
  aicommit2 config set topP=0.2
388
390
  ```
389
391
 
392
+ ##### codeReview
393
+
394
+ Default: `false`
395
+
396
+ The `codeReview` parameter determines whether to include an automated code review in the process.
397
+
398
+ ```sh
399
+ aicommit2 config set codeReview=true
400
+ ```
401
+
402
+ > NOTE: When enabled, aicommit2 will perform a code review before generating commit messages.
403
+
404
+ <img src="https://github.com/tak-bro/aicommit2/blob/main/img/code_review.gif?raw=true" alt="CODE_REVIEW" />
405
+
406
+ ⚠️ **CAUTION**
407
+
408
+ - The `codeReview` feature is currently experimental.
409
+ - This feature performs a code review before generating commit messages.
410
+ - Using this feature will significantly increase the overall processing time.
411
+ - It may significantly impact performance and cost.
412
+ - **The code review process consumes a large number of tokens, due to the lack of caching for git diff.**
413
+
414
+ ##### codeReviewPromptPath
415
+ - Allow users to specify a custom file path for code review
416
+
417
+ ```sh
418
+ aicommit2 config set codeReviewPromptPath="/path/to/user/prompt.txt"
419
+ ```
420
+
421
+
390
422
  ## Available General Settings by Model
391
- | | timeout | temperature | maxTokens | topP |
392
- |:--------------------:|:-------:|:-----------:|:---------:|:----:|
393
- | **OpenAI** | ✓ | ✓ | ✓ | ✓ |
394
- | **Anthropic Claude** | | ✓ | ✓ | |
395
- | **Gemini** | | ✓ | ✓ | |
396
- | **Mistral AI** | ✓ | ✓ | ✓ | ✓ |
397
- | **Codestral** | ✓ | ✓ | ✓ | ✓ |
398
- | **Cohere** | | ✓ | ✓ | |
399
- | **Groq** | ✓ | ✓ | ✓ | ✓ |
400
- | **Perplexity** | ✓ | ✓ | ✓ | ✓ |
401
- | **DeepSeek** | ✓ | ✓ | ✓ | ✓ |
402
- | **Huggingface** | | | | |
403
- | **Ollama** | ✓ | ✓ | | |
423
+ | | timeout | temperature | maxTokens | topP |
424
+ |:--------------------:|:-------:|:-----------:|:---------:|:------:|
425
+ | **OpenAI** | ✓ | ✓ | ✓ | ✓ |
426
+ | **Anthropic Claude** | | ✓ | ✓ | ✓ |
427
+ | **Gemini** | | ✓ | ✓ | ✓ |
428
+ | **Mistral AI** | ✓ | ✓ | ✓ | ✓ |
429
+ | **Codestral** | ✓ | ✓ | ✓ | ✓ |
430
+ | **Cohere** | | ✓ | ✓ | ✓ |
431
+ | **Groq** | ✓ | ✓ | ✓ | ✓ |
432
+ | **Perplexity** | ✓ | ✓ | ✓ | ✓ |
433
+ | **DeepSeek** | ✓ | ✓ | ✓ | ✓ |
434
+ | **Huggingface** | | | | |
435
+ | **Ollama** | ✓ | ✓ | | ✓ |
404
436
 
405
437
  > All AI support the following options in General Settings.
406
- > - systemPrompt, systemPromptPath, exclude, type, locale, generate, logging, ignoreBody, maxLength
438
+ > - systemPrompt, systemPromptPath, codeReview, codeReviewPromptPath, exclude, type, locale, generate, logging, includeBody, maxLength
407
439
 
408
440
  ## Model-Specific Settings
409
441
 
@@ -455,7 +487,7 @@ The OpenAI Path.
455
487
 
456
488
  ##### OPENAI.topP
457
489
 
458
- Default: `1`
490
+ Default: `0.9`
459
491
 
460
492
  The `top_p` parameter selects tokens whose combined probability meets a threshold. Please see [detail](https://platform.openai.com/docs/api-reference/chat/create#chat-create-top_p).
461
493
 
@@ -465,96 +497,36 @@ aicommit2 config set OPENAI.topP=0.2
465
497
 
466
498
  > NOTE: If `topP` is less than 0, it does not deliver the `top_p` parameter to the request.
467
499
 
468
- ### Ollama
469
-
470
- | Setting | Description | Default |
471
- |--------------------|----------------------------------------------|------------------------|
472
- | `model` | Model(s) to use (comma-separated list) | - |
473
- | `host` | Ollama host URL | http://localhost:11434 |
474
- | `timeout` | Request timeout (milliseconds) | 100_000 (100sec) |
475
-
476
- ##### OLLAMA.model
477
-
478
- The Ollama Model. Please see [a list of models available](https://ollama.com/library)
479
-
480
- ```sh
481
- aicommit2 config set OLLAMA.model="llama3.1"
482
- aicommit2 config set OLLAMA.model="llama3,codellama" # for multiple models
483
-
484
- aicommit2 config add OLLAMA.model="gemma2" # Only Ollama.model can be added.
485
- ```
486
-
487
- > OLLAMA.model is **string array** type to support multiple Ollama. Please see [this section](#loading-multiple-ollama-models).
488
-
489
- ##### OLLAMA.host
490
-
491
- Default: `http://localhost:11434`
492
-
493
- The Ollama host
494
-
495
- ```sh
496
- aicommit2 config set OLLAMA.host=<host>
497
- ```
498
-
499
- ##### OLLAMA.timeout
500
-
501
- Default: `100_000` (100 seconds)
502
-
503
- Request timeout for the Ollama.
504
-
505
- ```sh
506
- aicommit2 config set OLLAMA.timeout=<timeout>
507
- ```
508
-
509
- ##### Unsupported Options
510
-
511
- Ollama does not support the following options in General Settings.
512
-
513
- - maxTokens
514
- - topP
515
-
516
- ### HuggingFace
517
-
518
- | Setting | Description | Default |
519
- |--------------------|----------------------------|----------------------------------------|
520
- | `cookie` | Authentication cookie | - |
521
- | `model` | Model to use | `CohereForAI/c4ai-command-r-plus` |
500
+ ### Anthropic
522
501
 
523
- ##### HUGGINGFACE.cookie
502
+ | Setting | Description | Default |
503
+ |-------------|----------------|---------------------------|
504
+ | `key` | API key | - |
505
+ | `model` | Model to use | `claude-3-haiku-20240307` |
524
506
 
525
- The [Huggingface Chat](https://huggingface.co/chat/) Cookie. Please check [how to get cookie](https://github.com/tak-bro/aicommit2?tab=readme-ov-file#how-to-get-cookieunofficial-api)
507
+ ##### ANTHROPIC.key
526
508
 
527
- ```sh
528
- # Please be cautious of Escape characters(\", \') in browser cookie string
529
- aicommit2 config set HUGGINGFACE.cookie="your-cooke"
530
- ```
509
+ The Anthropic API key. To get started with Anthropic Claude, request access to their API at [anthropic.com/earlyaccess](https://www.anthropic.com/earlyaccess).
531
510
 
532
- ##### HUGGINGFACE.model
511
+ ##### ANTHROPIC.model
533
512
 
534
- Default: `CohereForAI/c4ai-command-r-plus`
513
+ Default: `claude-3-haiku-20240307`
535
514
 
536
515
  Supported:
537
- - `CohereForAI/c4ai-command-r-plus`
538
- - `meta-llama/Meta-Llama-3-70B-Instruct`
539
- - `HuggingFaceH4/zephyr-orpo-141b-A35b-v0.1`
540
- - `mistralai/Mixtral-8x7B-Instruct-v0.1`
541
- - `NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO`
542
- - `01-ai/Yi-1.5-34B-Chat`
543
- - `mistralai/Mistral-7B-Instruct-v0.2`
544
- - `microsoft/Phi-3-mini-4k-instruct`
516
+ - `claude-3-haiku-20240307`
517
+ - `claude-3-sonnet-20240229`
518
+ - `claude-3-opus-20240229`
519
+ - `claude-3-5-sonnet-20240620`
545
520
 
546
521
  ```sh
547
- aicommit2 config set HUGGINGFACE.model="mistralai/Mistral-7B-Instruct-v0.2"
522
+ aicommit2 config set ANTHROPIC.model="claude-3-5-sonnet-20240620"
548
523
  ```
549
524
 
550
525
  ##### Unsupported Options
551
526
 
552
- Huggingface does not support the following options in General Settings.
527
+ Anthropic does not support the following options in General Settings.
553
528
 
554
- - maxTokens
555
529
  - timeout
556
- - temperature
557
- - topP
558
530
 
559
531
  ### Gemini
560
532
 
@@ -589,39 +561,6 @@ aicommit2 config set GEMINI.model="gemini-1.5-pro-exp-0801"
589
561
  Gemini does not support the following options in General Settings.
590
562
 
591
563
  - timeout
592
- - topP
593
-
594
- ### Anthropic
595
-
596
- | Setting | Description | Default |
597
- |-------------|----------------|---------------------------|
598
- | `key` | API key | - |
599
- | `model` | Model to use | `claude-3-haiku-20240307` |
600
-
601
- ##### ANTHROPIC.key
602
-
603
- The Anthropic API key. To get started with Anthropic Claude, request access to their API at [anthropic.com/earlyaccess](https://www.anthropic.com/earlyaccess).
604
-
605
- ##### ANTHROPIC.model
606
-
607
- Default: `claude-3-haiku-20240307`
608
-
609
- Supported:
610
- - `claude-3-haiku-20240307`
611
- - `claude-3-sonnet-20240229`
612
- - `claude-3-opus-20240229`
613
- - `claude-3-5-sonnet-20240620`
614
-
615
- ```sh
616
- aicommit2 config set ANTHROPIC.model="claude-3-5-sonnet-20240620"
617
- ```
618
-
619
- ##### Unsupported Options
620
-
621
- Anthropic does not support the following options in General Settings.
622
-
623
- - timeout
624
- - topP
625
564
 
626
565
  ### Mistral
627
566
 
@@ -677,7 +616,7 @@ Supported:
677
616
  aicommit2 config set CODESTRAL.model="codestral-2405"
678
617
  ```
679
618
 
680
- #### Cohere
619
+ ### Cohere
681
620
 
682
621
  | Setting | Description | Default |
683
622
  |--------------------|--------------|-------------|
@@ -707,7 +646,6 @@ aicommit2 config set COHERE.model="command-nightly"
707
646
  Cohere does not support the following options in General Settings.
708
647
 
709
648
  - timeout
710
- - topP
711
649
 
712
650
  ### Groq
713
651
 
@@ -794,6 +732,96 @@ Supported:
794
732
  aicommit2 config set DEEPSEEK.model="deepseek-chat"
795
733
  ```
796
734
 
735
+ ### HuggingFace
736
+
737
+ | Setting | Description | Default |
738
+ |--------------------|----------------------------|----------------------------------------|
739
+ | `cookie` | Authentication cookie | - |
740
+ | `model` | Model to use | `CohereForAI/c4ai-command-r-plus` |
741
+
742
+ ##### HUGGINGFACE.cookie
743
+
744
+ The [Huggingface Chat](https://huggingface.co/chat/) Cookie. Please check [how to get cookie](https://github.com/tak-bro/aicommit2?tab=readme-ov-file#how-to-get-cookieunofficial-api)
745
+
746
+ ```sh
747
+ # Please be cautious of Escape characters(\", \') in browser cookie string
748
+ aicommit2 config set HUGGINGFACE.cookie="your-cooke"
749
+ ```
750
+
751
+ ##### HUGGINGFACE.model
752
+
753
+ Default: `CohereForAI/c4ai-command-r-plus`
754
+
755
+ Supported:
756
+ - `CohereForAI/c4ai-command-r-plus`
757
+ - `meta-llama/Meta-Llama-3-70B-Instruct`
758
+ - `HuggingFaceH4/zephyr-orpo-141b-A35b-v0.1`
759
+ - `mistralai/Mixtral-8x7B-Instruct-v0.1`
760
+ - `NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO`
761
+ - `01-ai/Yi-1.5-34B-Chat`
762
+ - `mistralai/Mistral-7B-Instruct-v0.2`
763
+ - `microsoft/Phi-3-mini-4k-instruct`
764
+
765
+ ```sh
766
+ aicommit2 config set HUGGINGFACE.model="mistralai/Mistral-7B-Instruct-v0.2"
767
+ ```
768
+
769
+ ##### Unsupported Options
770
+
771
+ Huggingface does not support the following options in General Settings.
772
+
773
+ - maxTokens
774
+ - timeout
775
+ - temperature
776
+ - topP
777
+
778
+ ### Ollama
779
+
780
+ | Setting | Description | Default |
781
+ |--------------------|----------------------------------------------|------------------------|
782
+ | `model` | Model(s) to use (comma-separated list) | - |
783
+ | `host` | Ollama host URL | http://localhost:11434 |
784
+ | `timeout` | Request timeout (milliseconds) | 100_000 (100sec) |
785
+
786
+ ##### OLLAMA.model
787
+
788
+ The Ollama Model. Please see [a list of models available](https://ollama.com/library)
789
+
790
+ ```sh
791
+ aicommit2 config set OLLAMA.model="llama3.1"
792
+ aicommit2 config set OLLAMA.model="llama3,codellama" # for multiple models
793
+
794
+ aicommit2 config add OLLAMA.model="gemma2" # Only Ollama.model can be added.
795
+ ```
796
+
797
+ > OLLAMA.model is **string array** type to support multiple Ollama. Please see [this section](#loading-multiple-ollama-models).
798
+
799
+ ##### OLLAMA.host
800
+
801
+ Default: `http://localhost:11434`
802
+
803
+ The Ollama host
804
+
805
+ ```sh
806
+ aicommit2 config set OLLAMA.host=<host>
807
+ ```
808
+
809
+ ##### OLLAMA.timeout
810
+
811
+ Default: `100_000` (100 seconds)
812
+
813
+ Request timeout for the Ollama.
814
+
815
+ ```sh
816
+ aicommit2 config set OLLAMA.timeout=<timeout>
817
+ ```
818
+
819
+ ##### Unsupported Options
820
+
821
+ Ollama does not support the following options in General Settings.
822
+
823
+ - maxTokens
824
+
797
825
  ## Upgrading
798
826
 
799
827
  Check the installed version with:
@@ -834,7 +862,7 @@ Use curly braces `{}` to denote these placeholders for options. The following pl
834
862
  - [{type}](#type): The type of the commit message (**conventional** or **gitmoji**)
835
863
  - [{generate}](#generate): The number of commit messages to generate (**number**)
836
864
 
837
- ### Example Template
865
+ #### Example Template
838
866
 
839
867
  Here's an example of how your custom template might look:
840
868
 
@@ -849,16 +877,16 @@ Remember to follow these guidelines:
849
877
  3. Explain the 'why' behind the change
850
878
  ```
851
879
 
852
- ### **Appended Text**
880
+ #### **Appended Text**
853
881
 
854
- Please note that the following text will **always** be appended to the end of your custom prompt:
882
+ Please note that the following text will **ALWAYS** be appended to the end of your custom prompt:
855
883
 
856
884
  ```
857
- Provide your response as a JSON array containing exactly 1 object, each with the following keys:
858
- - "subject": The main commit message using the conventional style. It should be a concise summary of the changes.
885
+ Lastly, Provide your response as a JSON array containing exactly {generate} object, each with the following keys:
886
+ - "subject": The main commit message using the {type} style. It should be a concise summary of the changes.
859
887
  - "body": An optional detailed explanation of the changes. If not needed, use an empty string.
860
888
  - "footer": An optional footer for metadata like BREAKING CHANGES. If not needed, use an empty string.
861
- The array must always contain 1 element, no more and no less.
889
+ The array must always contain {generate} element, no more and no less.
862
890
  Example response format:
863
891
  [
864
892
  {
@@ -867,14 +895,12 @@ Example response format:
867
895
  "footer": ""
868
896
  }
869
897
  ]
870
- Ensure you generate exactly 1 commit message, even if it requires creating slightly varied versions for similar changes.
898
+ Ensure you generate exactly {generate} commit message, even if it requires creating slightly varied versions for similar changes.
871
899
  The response should be valid JSON that can be parsed without errors.
872
900
  ```
873
901
 
874
902
  This ensures that the output is consistently formatted as a JSON array, regardless of the custom template used.
875
903
 
876
- > NOTE: The template may vary depending on the generate and commit message type.
877
-
878
904
  ## Loading Multiple Ollama Models
879
905
 
880
906
  <img src="https://github.com/tak-bro/aicommit2/blob/main/img/ollama_parallel.gif?raw=true" alt="OLLAMA_PARALLEL" />