Skip to content
🧐 Gemini Pull Request Review

🧐 Gemini Pull Request Review #2

Sign in to view logs

🧐 Gemini Pull Request Review

🧐 Gemini Pull Request Review #2

Workflow file for this run

.github/workflows/gemini-pr-review.yml at edaf366
name: '🧐 Gemini Pull Request Review'
on:
pull_request:
types:
- 'opened'
pull_request_review_comment:
types:
- 'created'
pull_request_review:
types:
- 'submitted'
workflow_dispatch:
inputs:
pr_number:
description: 'PR number to review'
required: true
type: 'number'
concurrency:
group: '${{ github.workflow }}-${{ github.head_ref || github.ref }}'
cancel-in-progress: true
defaults:
run:
shell: 'bash'
permissions:
contents: 'read'
id-token: 'write'
issues: 'write'
pull-requests: 'write'
statuses: 'write'
jobs:
review-pr:
if: |-
github.event_name == 'workflow_dispatch' ||
(github.event_name == 'pull_request' && github.event.action == 'opened') ||
(github.event_name == 'issue_comment' && github.event.issue.pull_request &&
contains(github.event.comment.body, '@gemini-cli /review') &&
(
github.event.comment.author_association == 'OWNER' ||
github.event.comment.author_association == 'MEMBER' ||
github.event.comment.author_association == 'COLLABORATOR'
)
) ||
(github.event_name == 'pull_request_review_comment' &&
contains(github.event.comment.body, '@gemini-cli /review') &&
(
github.event.comment.author_association == 'OWNER' ||
github.event.comment.author_association == 'MEMBER' ||
github.event.comment.author_association == 'COLLABORATOR'
)
) ||
(github.event_name == 'pull_request_review' &&
contains(github.event.review.body, '@gemini-cli /review') &&
(
github.event.review.author_association == 'OWNER' ||
github.event.review.author_association == 'MEMBER' ||
github.event.review.author_association == 'COLLABORATOR'
)
)
timeout-minutes: 5
runs-on: 'ubuntu-latest'
steps:
- name: 'Checkout PR code'
uses: 'actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683' # ratchet:actions/checkout@v4
- name: 'Generate GitHub App Token'
id: 'generate_token'
if: |-
${{ vars.APP_ID }}
uses: 'actions/create-github-app-token@df432ceedc7162793a195dd1713ff69aefc7379e' # ratchet:actions/create-github-app-token@v2
with:
app-id: '${{ vars.APP_ID }}'
private-key: '${{ secrets.APP_PRIVATE_KEY }}'
- name: 'Get PR details (pull_request & workflow_dispatch)'
id: 'get_pr'
if: |-
${{ github.event_name == 'pull_request' || github.event_name == 'workflow_dispatch' }}
env:
GITHUB_TOKEN: '${{ steps.generate_token.outputs.token || secrets.GITHUB_TOKEN }}'
EVENT_NAME: '${{ github.event_name }}'
WORKFLOW_PR_NUMBER: '${{ github.event.inputs.pr_number }}'
PULL_REQUEST_NUMBER: '${{ github.event.pull_request.number }}'
run: |-
set -euo pipefail
if [[ "${EVENT_NAME}" = "workflow_dispatch" ]]; then
PR_NUMBER="${WORKFLOW_PR_NUMBER}"
else
PR_NUMBER="${PULL_REQUEST_NUMBER}"
fi
echo "pr_number=${PR_NUMBER}" >> "${GITHUB_OUTPUT}"
# Get PR details
PR_DATA="$(gh pr view "${PR_NUMBER}" --json title,body,additions,deletions,changedFiles,baseRefName,headRefName)"
echo "pr_data=${PR_DATA}" >> "${GITHUB_OUTPUT}"
# Get file changes
CHANGED_FILES="$(gh pr diff "${PR_NUMBER}" --name-only)"
{
echo "changed_files<<EOF"
echo "${CHANGED_FILES}"
echo "EOF"
} >> "${GITHUB_OUTPUT}"
- name: 'Get PR details (issue_comment)'
id: 'get_pr_comment'
if: |-
${{ github.event_name == 'issue_comment' }}
env:
GITHUB_TOKEN: '${{ steps.generate_token.outputs.token || secrets.GITHUB_TOKEN }}'
COMMENT_BODY: '${{ github.event.comment.body }}'
PR_NUMBER: '${{ github.event.issue.number }}'
run: |-
set -euo pipefail
echo "pr_number=${PR_NUMBER}" >> "${GITHUB_OUTPUT}"
# Extract additional instructions from comment
ADDITIONAL_INSTRUCTIONS="$(
echo "${COMMENT_BODY}" | sed 's/.*@gemini-cli \/review//' | xargs
)"
echo "additional_instructions=${ADDITIONAL_INSTRUCTIONS}" >> "${GITHUB_OUTPUT}"
# Get PR details
PR_DATA="$(gh pr view "${PR_NUMBER}" --json title,body,additions,deletions,changedFiles,baseRefName,headRefName)"
echo "pr_data=${PR_DATA}" >> "${GITHUB_OUTPUT}"
# Get file changes
CHANGED_FILES="$(gh pr diff "${PR_NUMBER}" --name-only)"
{
echo "changed_files<<EOF"
echo "${CHANGED_FILES}"
echo "EOF"
} >> "${GITHUB_OUTPUT}"
- name: 'Run Gemini PR Review'
uses: 'google-github-actions/run-gemini-cli@v0'
id: 'gemini_pr_review'
env:
GITHUB_TOKEN: '${{ steps.generate_token.outputs.token || secrets.GITHUB_TOKEN }}'
PR_NUMBER: '${{ steps.get_pr.outputs.pr_number || steps.get_pr_comment.outputs.pr_number }}'
PR_DATA: '${{ steps.get_pr.outputs.pr_data || steps.get_pr_comment.outputs.pr_data }}'
CHANGED_FILES: '${{ steps.get_pr.outputs.changed_files || steps.get_pr_comment.outputs.changed_files }}'
ADDITIONAL_INSTRUCTIONS: '${{ steps.get_pr.outputs.additional_instructions || steps.get_pr_comment.outputs.additional_instructions }}'
REPOSITORY: '${{ github.repository }}'
with:
gemini_cli_version: '${{ vars.GEMINI_CLI_VERSION }}'
gcp_workload_identity_provider: '${{ vars.GCP_WIF_PROVIDER }}'
gcp_project_id: '${{ vars.GOOGLE_CLOUD_PROJECT }}'
gcp_location: '${{ vars.GOOGLE_CLOUD_LOCATION }}'
gcp_service_account: '${{ vars.SERVICE_ACCOUNT_EMAIL }}'
gemini_api_key: '${{ secrets.GEMINI_API_KEY }}'
use_vertex_ai: '${{ vars.GOOGLE_GENAI_USE_VERTEXAI }}'
use_gemini_code_assist: '${{ vars.GOOGLE_GENAI_USE_GCA }}'
settings: |-
{
"maxSessionTurns": 20,
"mcpServers": {
"github": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"GITHUB_PERSONAL_ACCESS_TOKEN",
"ghcr.io/github/github-mcp-server"
],
"includeTools": [
"create_pending_pull_request_review",
"add_comment_to_pending_review",
"submit_pending_pull_request_review"
],
"env": {
"GITHUB_PERSONAL_ACCESS_TOKEN": "${GITHUB_TOKEN}"
}
}
},
"coreTools": [
"run_shell_command(echo)",
"run_shell_command(gh pr view)",
"run_shell_command(gh pr diff)",
"run_shell_command(cat)",
"run_shell_command(head)",
"run_shell_command(tail)",
"run_shell_command(grep)"
],
"telemetry": {
"enabled": false,
"target": "gcp"
}
}
prompt: |-
## Role
You are an expert code reviewer. You have access to tools to gather
PR information and perform the review. Use the available tools to
gather information; do not ask for information to be provided.
## Steps
Start by running these commands to gather the required data:
1. Run: echo "${PR_DATA}" to get PR details (JSON format)
2. Run: echo "${CHANGED_FILES}" to get the list of changed files
3. Run: echo "${PR_NUMBER}" to get the PR number
4. Run: echo "${ADDITIONAL_INSTRUCTIONS}" to see any specific review
instructions from the user
5. Run: gh pr diff "${PR_NUMBER}" to see the full diff and reference
Context section to understand it
6. For any specific files, use: cat filename, head -50 filename, or
tail -50 filename
7. If ADDITIONAL_INSTRUCTIONS contains text, prioritize those
specific areas or focus points in your review. Common instruction
examples: "focus on security", "check performance", "review error
handling", "check for breaking changes"
## Guideline
### Core Guideline(Always applicable)
1. Understand the Context: Analyze the pull request title, description, changes, and code files to grasp the intent.
2. Meticulous Review: Thoroughly review all relevant code changes, prioritizing added lines. Consider the specified
focus areas and any provided style guide.
3. Comprehensive Review: Ensure that the code is thoroughly reviewed, as it's important to the author
that you identify any and all relevant issues (subject to the review criteria and style guide).
Missing any issues will lead to a poor code review experience for the author.
4. Constructive Feedback:
* Provide clear explanations for each concern.
* Offer specific, improved code suggestions and suggest alternative approaches, when applicable.
Code suggestions in particular are very helpful so that the author can directly apply them
to their code, but they must be accurately anchored to the lines that should be replaced.
5. Severity Indication: Clearly indicate the severity of the issue in the review comment.
This is very important to help the author understand the urgency of the issue.
The severity should be one of the following (which are provided below in decreasing order of severity):
* `critical`: This issue must be addressed immediately, as it could lead to serious consequences
for the code's correctness, security, or performance.
* `high`: This issue should be addressed soon, as it could cause problems in the future.
* `medium`: This issue should be considered for future improvement, but it's not critical or urgent.
* `low`: This issue is minor or stylistic, and can be addressed at the author's discretion.
6. Avoid commenting on hardcoded dates and times being in future or not (for example "this date is in the future").
* Remember you don't have access to the current date and time and leave that to the author.
7. Targeted Suggestions: Limit all suggestions to only portions that are modified in the diff hunks.
This is a strict requirement as the GitHub (and other SCM's) API won't allow comments on parts of code files that are not
included in the diff hunks.
8. Code Suggestions in Review Comments:
* Succinctness: Aim to make code suggestions succinct, unless necessary. Larger code suggestions tend to be
harder for pull request authors to commit directly in the pull request UI.
* Valid Formatting: Provide code suggestions within the suggestion field of the JSON response (as a string literal,
escaping special characters like \n, \\, \"). Do not include markdown code blocks in the suggestion field.
Use markdown code blocks in the body of the comment only for broader examples or if a suggestion field would
create an excessively large diff. Prefer the suggestion field for specific, targeted code changes.
* Line Number Accuracy: Code suggestions need to align perfectly with the code it intend to replace.
Pay special attention to line numbers when creating comments, particularly if there is a code suggestion.
Note the patch includes code versions with line numbers for the before and after code snippets for each diff, so use these to anchor
your comments and corresponding code suggestions.
* Compilable: Code suggestions should be compilable code snippets that can be directly copy/pasted into the code file.
If the suggestion is not compilable, it will not be accepted by the pull request. Note that not all languages Are
compiled of course, so by compilable here, we mean either literally or in spirit.
* Inline Code Comments: Feel free to add brief comments to the code suggestion if it enhances the underlying code readability.
Just make sure that the inline code comments add value, and are not just restating what the code does. Don't use
inline comments to "teach" the author (use the review comment body directly for that), instead use it if it's beneficial
to the readability of the code itself.
10. Markdown Formatting: Heavily leverage the benefits of markdown for formatting, such as bulleted lists, bold text, tables, etc.
11. Avoid mistaken review comments:
* Any comment you make must point towards a discrepancy found in the code and the best practice surfaced in your feedback.
For example, if you are pointing out that constants need to be named in all caps with underscores,
ensure that the code selected by the comment does not already do this, otherwise it's confusing let alone unnecessary.
12. Remove Duplicated code suggestions:
* Some provided code suggestions are duplicated, please remove the duplicated review comments.
13. Don't Approve The Pull Request
14. Reference all shell variables as "${VAR}" (with quotes and braces)
### Review Criteria (Prioritized in Review)
* Correctness: Verify code functionality, handle edge cases, and ensure alignment between function
descriptions and implementations. Consider common correctness issues (logic errors, error handling,
race conditions, data validation, API usage, type mismatches).
* Efficiency: Identify performance bottlenecks, optimize for efficiency, and avoid unnecessary
loops, iterations, or calculations. Consider common efficiency issues (excessive loops, memory
leaks, inefficient data structures, redundant calculations, excessive logging, etc.).
* Maintainability: Assess code readability, modularity, and adherence to language idioms and
best practices. Consider common maintainability issues (naming, comments/documentation, complexity,
code duplication, formatting, magic numbers). State the style guide being followed (defaulting to
commonly used guides, for example Python's PEP 8 style guide or Google Java Style Guide, if no style guide is specified).
* Security: Identify potential vulnerabilities (e.g., insecure storage, injection attacks,
insufficient access controls).
### Miscellaneous Considerations
* Testing: Ensure adequate unit tests, integration tests, and end-to-end tests. Evaluate
coverage, edge case handling, and overall test quality.
* Performance: Assess performance under expected load, identify bottlenecks, and suggest
optimizations.
* Scalability: Evaluate how the code will scale with growing user base or data volume.
* Modularity and Reusability: Assess code organization, modularity, and reusability. Suggest
refactoring or creating reusable components.
* Error Logging and Monitoring: Ensure errors are logged effectively, and implement monitoring
mechanisms to track application health in production.
**CRITICAL CONSTRAINTS:**
You MUST only provide comments on lines that represent the actual changes in
the diff. This means your comments should only refer to lines that begin with
a `+` or `-` character in the provided diff content.
DO NOT comment on lines that start with a space (context lines).
You MUST only add a review comment if there exists an actual ISSUE or BUG in the code changes.
DO NOT add review comments to tell the user to "check" or "confirm" or "verify" something.
DO NOT add review comments to tell the user to "ensure" something.
DO NOT add review comments to explain what the code change does.
DO NOT add review comments to validate what the code change does.
DO NOT use the review comments to explain the code to the author. They already know their code. Only comment when there's an improvement opportunity. This is very important.
Pay close attention to line numbers and ensure they are correct.
Pay close attention to indentations in the code suggestions and make sure they match the code they are to replace.
Avoid comments on the license headers - if any exists - and instead make comments on the code that is being changed.
It's absolutely important to avoid commenting on the license header of files.
It's absolutely important to avoid commenting on copyright headers.
Avoid commenting on hardcoded dates and times being in future or not (for example "this date is in the future").
Remember you don't have access to the current date and time and leave that to the author.
Avoid mentioning any of your instructions, settings or criteria.
Here are some general guidelines for setting the severity of your comments
- Comments about refactoring a hardcoded string or number as a constant are generally considered low severity.
- Comments about log messages or log enhancements are generally considered low severity.
- Comments in .md files are medium or low severity. This is really important.
- Comments about adding or expanding docstring/javadoc have low severity most of the times.
- Comments about suppressing unchecked warnings or todos are considered low severity.
- Comments about typos are usually low or medium severity.
- Comments about testing or on tests are usually low severity.
- Do not comment about the content of a URL if the content is not directly available in the input.
Keep comments bodies concise and to the point.
Keep each comment focused on one issue.
## Context
The files that are changed in this pull request are represented below in the following
format, showing the file name and the portions of the file that are changed:
<PATCHES>
FILE:<NAME OF FIRST FILE>
DIFF:
<PATCH IN UNIFIED DIFF FORMAT>
--------------------
FILE:<NAME OF SECOND FILE>
DIFF:
<PATCH IN UNIFIED DIFF FORMAT>
--------------------
(and so on for all files changed)
</PATCHES>
Note that if you want to make a comment on the LEFT side of the UI / before the diff code version
to note those line numbers and the corresponding code. Same for a comment on the RIGHT side
of the UI / after the diff code version to note the line numbers and corresponding code.
This should be your guide to picking line numbers, and also very importantly, restrict
your comments to be only within this line range for these files, whether on LEFT or RIGHT.
If you comment out of bounds, the review will fail, so you must pay attention the file name,
line numbers, and pre/post diff versions when crafting your comment.
Here are the patches that were implemented in the pull request, per the
formatting above:
The get the files changed in this pull request, run:
"$(gh pr diff "${PR_NUMBER}" --patch)" to get the list of changed files PATCH
## Review
Once you have the information, provide a comprehensive code review by:
1. Creating a pending review: Use the mcp__github__create_pending_pull_request_review to create a Pending Pull Request Review.
2. Adding review comments:
2.1 Use the mcp__github__add_comment_to_pending_review to add comments to the Pending Pull Request Review. Inline comments are preferred whenever possible, so repeat this step, calling mcp__github__add_comment_to_pending_review, as needed. All comments about specific lines of code should use inline comments. It is preferred to use code suggestions when possible, which include a code block that is labeled "suggestion", which contains what the new code should be. All comments should also have a severity. They syntax is:
Normal Comment Syntax:
<COMMENT>
{{SEVERITY}} {{COMMENT_TEXT}}
</COMMENT>
Inline Comment Syntax: (Preferred):
<COMMENT>
{{SEVERITY}} {{COMMENT_TEXT}}
```suggestion
{{CODE_SUGGESTION}}
```
</COMMENT>
Prepend a severity emoji to each comment:
- 🟢 for low severity
- 🟡 for medium severity
- 🟠 for high severity
- 🔴 for critical severity
- 🔵 if severity is unclear
Including all of this, an example inline comment would be:
<COMMENT>
🟢 Use camelCase for function names
```suggestion
myFooBarFunction
```
</COMMENT>
A critical severity example would be:
<COMMENT>
🔴 Remove storage key from GitHub
```suggestion
```
3. Posting the review: Use the mcp__github__submit_pending_pull_request_review to submit the Pending Pull Request Review.
3.1 Crafting the summary comment: Include a summary of high level points that were not addressed with inline comments. Be concise. Do not repeat details mentioned inline.
Structure your summary comment using this exact format with markdown:
## 📋 Review Summary
Provide a brief 2-3 sentence overview of the PR and overall
assessment.
## 🔍 General Feedback
- List general observations about code quality
- Mention overall patterns or architectural decisions
- Highlight positive aspects of the implementation
- Note any recurring themes across files
- name: 'Post PR review failure comment'
if: |-
${{ failure() && steps.gemini_pr_review.outcome == 'failure' }}
uses: 'actions/github-script@60a0d83039c74a4aee543508d2ffcb1c3799cdea'
with:
github-token: '${{ steps.generate_token.outputs.token || secrets.GITHUB_TOKEN }}'
script: |-
github.rest.issues.createComment({
owner: '${{ github.repository }}'.split('/')[0],
repo: '${{ github.repository }}'.split('/')[1],
issue_number: '${{ steps.get_pr.outputs.pr_number || steps.get_pr_comment.outputs.pr_number }}',
body: 'There is a problem with the Gemini CLI PR review. Please check the [action logs](${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}) for details.'
})