open-code-review

Name: open-code-review
Author: alibaba

2.3kinstalls751forks

How to install open-code-review?

alibaba/open-code-review/open-code-review

$npx -y skills add alibaba/open-code-review --skill open-code-review

Installs into the current project.

Summary

Performs AI-powered code review on Git changes using the ocr CLI from alibaba/open-code-review. Use when the user asks to review code, review a pull request, review staged/unstaged changes, review a commit, or compare branches for code quality issues. Produces line-level review

Files · 1

View on GitHub

SKILL.md

1	# Open Code Review
2
3	This Codex plugin skill intentionally mirrors the canonical skill at
4	`skills/open-code-review/SKILL.md`. Keep both files synchronized when updating
5	OCR agent instructions; a symlink is avoided because plugin installs may only
6	materialize the plugin subtree.
7
8	A skill for invoking [open-code-review](https://github.com/alibaba/open-code-review) (`ocr`) — an open-source AI code review CLI that reads Git diffs and generates structured, line-level review comments.
9
10	## Prerequisites check
11
12	Before starting a review, verify the environment:
13
14	```bash
15	# 1. Check the CLI is installed
16	which ocr \|\| echo "NOT INSTALLED"
17
18	# 2. Verify LLM connectivity
19	ocr llm test
20	```
21
22	If `ocr` is not installed, install it first:
23
24	```bash
25	npm install -g @alibaba-group/open-code-review
26	```
27
28	If `ocr llm test` fails, the user must configure an LLM. Guide them with one of these options:
29
30	Option A — Environment variables (highest priority, recommended for CI):
31
32	```bash
33	export OCR_LLM_URL=https://api.anthropic.com/v1/messages
34	export OCR_LLM_TOKEN=<api-key>
35	export OCR_LLM_MODEL=claude-opus-4-6
36	export OCR_USE_ANTHROPIC=true
37	```
38
39	Option B — Persistent config:
40
41	```bash
42	ocr config set llm.url https://api.anthropic.com/v1/messages
43	ocr config set llm.auth_token <api-key>
44	ocr config set llm.model claude-opus-4-6
45	ocr config set llm.use_anthropic true
46	```
47
48	Stop here and ask the user to provide credentials — never invent or hardcode API keys.
49
50	## Workflow
51
52	### Step 1: Gather Business Context
53
54	Analyze the review target (commits, branch, or changes) to extract concise business context. Pass this context via `--background` to improve review quality.
55
56	### Step 2: Run Code Review
57
58	Run the OCR command with appropriate flags. Always pass business context via `--background` when available:
59
60	```bash
61	ocr review --audience agent --background "business context here" [user-args]
62	```
63
64	Argument handling:
65
66	- Background context (RECOMMENDED): use `--background "context"` or `-b "context"` to provide business context for better review quality
67	- Default (no user arguments): reviews staged, unstaged, and untracked changes (workspace mode)
68	- Specific commit: use `--commit` or `-c` to review a single commit against its parent
69	- Branch comparison: use `--from <ref>` and `--to <ref>` to review diff between two refs
70	- Timeout: default timeout is 10 minutes per file; adjust with `--timeout <minutes>`
71	- Concurrency: default concurrency is 8 file workers; reduce with `--concurrency <n>` if rate limits are hit
72	- Preview mode: use `--preview` or `-p` to preview which files will be reviewed without running the LLM
73	- Installation: if `ocr` command is not found, install it by running `npm i -g @alibaba-group/open-code-review`
74
75	Common invocation patterns:
76
77	\| User says \| Command to run \|
78	\|-----------\|---------------\|
79	\| "review my changes" / "review the working copy" \| `ocr review --audience agent -b "context"` \|
80	\| "review this PR" / "review feature branch" \| `ocr review --audience agent -b "context" --from main --to <branch>` \|
81	\| "review commit abc123" \| `ocr review --audience agent -b "context" --commit abc123` \|
82	\| "what would be reviewed?" (dry-run) \| `ocr review --preview` \|
83
84	Output mode:
85
86	- Always use `--audience agent` to suppress progress UI and emit only the final summary
87
88	### Step 3: Classify and Report
89
90	For each comment from the review output, classify by priority and report all issues to the user:
91
92	- High: Obvious bugs, security issues, clear mistakes, or well-founded suggestions with precise fix proposals
93	- Medium: Reasonable concerns but context-dependent, style/performance suggestions, or fixes that require manual implementation
94	- Low: Likely false positives, lacking sufficient context, nitpicks, or meaningless suggestions
95
96	Report all comments grouped by priority level.
97
98	### Step 4: Fix
99
100	Before applying fixes, check whether the user requested automatic fixes:
101
102	- If the user explicitly requested "review and fix" or similar, proceed with automatic fixes
103	- If the user only requested "review" without fix intent, ask fo

People who liked this also viewed

6 picks

Type

open-code-review

2.3kinstalls751forks

View on GitHub

How to install open-code-review?

alibaba/open-code-review/open-code-review

$npx -y skills add alibaba/open-code-review --skill open-code-review

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Open Code Review
2
3	This Codex plugin skill intentionally mirrors the canonical skill at
4	`skills/open-code-review/SKILL.md`. Keep both files synchronized when updating
5	OCR agent instructions; a symlink is avoided because plugin installs may only
6	materialize the plugin subtree.
7
8	A skill for invoking [open-code-review](https://github.com/alibaba/open-code-review) (`ocr`) — an open-source AI code review CLI that reads Git diffs and generates structured, line-level review comments.
9
10	## Prerequisites check
11
12	Before starting a review, verify the environment:
13
14	```bash
15	# 1. Check the CLI is installed
16	which ocr \|\| echo "NOT INSTALLED"
17
18	# 2. Verify LLM connectivity
19	ocr llm test
20	```
21
22	If `ocr` is not installed, install it first:
23
24	```bash
25	npm install -g @alibaba-group/open-code-review
26	```
27
28	If `ocr llm test` fails, the user must configure an LLM. Guide them with one of these options:
29
30	Option A — Environment variables (highest priority, recommended for CI):
31
32	```bash
33	export OCR_LLM_URL=https://api.anthropic.com/v1/messages
34	export OCR_LLM_TOKEN=<api-key>
35	export OCR_LLM_MODEL=claude-opus-4-6
36	export OCR_USE_ANTHROPIC=true
37	```
38
39	Option B — Persistent config:
40
41	```bash
42	ocr config set llm.url https://api.anthropic.com/v1/messages
43	ocr config set llm.auth_token <api-key>
44	ocr config set llm.model claude-opus-4-6
45	ocr config set llm.use_anthropic true
46	```
47
48	Stop here and ask the user to provide credentials — never invent or hardcode API keys.
49
50	## Workflow
51
52	### Step 1: Gather Business Context
53
54	Analyze the review target (commits, branch, or changes) to extract concise business context. Pass this context via `--background` to improve review quality.
55
56	### Step 2: Run Code Review
57
58	Run the OCR command with appropriate flags. Always pass business context via `--background` when available:
59
60	```bash
61	ocr review --audience agent --background "business context here" [user-args]
62	```
63
64	Argument handling:
65
66	- Background context (RECOMMENDED): use `--background "context"` or `-b "context"` to provide business context for better review quality
67	- Default (no user arguments): reviews staged, unstaged, and untracked changes (workspace mode)
68	- Specific commit: use `--commit` or `-c` to review a single commit against its parent
69	- Branch comparison: use `--from <ref>` and `--to <ref>` to review diff between two refs
70	- Timeout: default timeout is 10 minutes per file; adjust with `--timeout <minutes>`
71	- Concurrency: default concurrency is 8 file workers; reduce with `--concurrency <n>` if rate limits are hit
72	- Preview mode: use `--preview` or `-p` to preview which files will be reviewed without running the LLM
73	- Installation: if `ocr` command is not found, install it by running `npm i -g @alibaba-group/open-code-review`
74
75	Common invocation patterns:
76
77	\| User says \| Command to run \|
78	\|-----------\|---------------\|
79	\| "review my changes" / "review the working copy" \| `ocr review --audience agent -b "context"` \|
80	\| "review this PR" / "review feature branch" \| `ocr review --audience agent -b "context" --from main --to <branch>` \|
81	\| "review commit abc123" \| `ocr review --audience agent -b "context" --commit abc123` \|
82	\| "what would be reviewed?" (dry-run) \| `ocr review --preview` \|
83
84	Output mode:
85
86	- Always use `--audience agent` to suppress progress UI and emit only the final summary
87
88	### Step 3: Classify and Report
89
90	For each comment from the review output, classify by priority and report all issues to the user:
91
92	- High: Obvious bugs, security issues, clear mistakes, or well-founded suggestions with precise fix proposals
93	- Medium: Reasonable concerns but context-dependent, style/performance suggestions, or fixes that require manual implementation
94	- Low: Likely false positives, lacking sufficient context, nitpicks, or meaningless suggestions
95
96	Report all comments grouped by priority level.
97
98	### Step 4: Fix
99
100	Before applying fixes, check whether the user requested automatic fixes:
101
102	- If the user explicitly requested "review and fix" or similar, proceed with automatic fixes
103	- If the user only requested "review" without fix intent, ask fo

People who liked this also viewed

6 picks

Type