GitHub Action
issue-metrics
This is a GitHub Action that searches for pull requests/issues/discussions in a repository and measures and reports on several metrics. The issues/pull requests/discussions to search for can be filtered by using a search query.
The metrics that are measured are:
Metric | Description |
---|---|
Time to first response | The time between when an issue/pull request/discussion is created and when the first comment or review is made. |
Time to close | The time between when an issue/pull request/discussion is created and when it is closed. |
Time to answer | (Discussions only) The time between when a discussion is created and when it is answered. |
Time in label | The time between when a label has a specific label applied to an issue/pull request/discussion and when it is removed. This requires the LABELS_TO_MEASURE env variable to be set. |
This action was developed by the GitHub OSPO for our own use and developed in a way that we could open source it that it might be useful to you as well! If you want to know more about how we use it, reach out in an issue in this repository.
To find syntax for search queries, check out the documentation on searching issues and pull requests or the documentation on searching discussions.
- As a maintainer, I want to see metrics for issues and pull requests on the repository I maintain in order to ensure I am giving them the proper amount of attention.
- As a first responder on a repository, I want to ensure that users are getting contact from me in a reasonable amount of time.
- As an OSPO, I want to see how many open source repository requests are open/closed, and metrics for how long it takes to get through the open source process.
- As a product development team, I want to see metrics around how long pull request reviews are taking, so that we can reflect on that data during retrospectives.
If you need support using this project or have questions about it, please open up an issue in this repository. Requests made directly to GitHub staff or support team will be redirected here to open an issue. GitHub SLA's and support/services contracts do not apply to this repository.
- Create a repository to host this GitHub Action or select an existing repository. This is easiest if it is the same repository as the one you want to measure metrics on.
- Select a best fit workflow file from the examples below.
- Copy that example into your repository (from step 1) and into the proper directory for GitHub Actions:
.github/workflows/
directory with the file extension.yml
(ie..github/workflows/issue-metrics.yml
) - Edit the values (
SEARCH_QUERY
,assignees
) from the sample workflow with your information. See the SEARCH_QUERY section for more details on options. - If you are running metrics on a repository other than the one where the workflow file is going to be, then update the value of
GH_TOKEN
. Do this by creating a GitHub API token with permissions to read the repo and write issues. Then take the value of the API token you just created, and create a repository secret where the name of the secret isGH_TOKEN
and the value of the secret the API token. Then finally update the workflow file to use that repository secret by changingGH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
toGH_TOKEN: ${{ secrets.GH_TOKEN }}
. The name of the secret can really be anything. It just needs to match between when you create the secret name and when you refer to it in the workflow file. - If you want the resulting issue with the metrics in it to appear in a different repository other than the one the workflow file runs in, update the line
token: ${{ secrets.GITHUB_TOKEN }}
with your own GitHub API token stored as a repository secret. This process is the same as described in the step above. More info on creating secrets can be found here. - Commit the workflow file to the default branch (often
master
ormain
) - Wait for the action to trigger based on the
schedule
entry or manually trigger the workflow as shown in the documentation.
Below are the allowed configuration options:
field | required | default | description |
---|---|---|---|
GH_TOKEN |
True | The GitHub Token used to scan the repository. Must have read access to all repository you are interested in scanning. | |
SEARCH_QUERY |
True | The query by which you can filter issues/prs which must contain a repo: , org: , owner: , or a user: entry. For discussions, include type:discussions in the query. |
|
LABELS_TO_MEASURE |
False | A comma separated list of labels to measure how much time the label is applied. If not provided, no labels durations will be measured. Not compatible with discussions at this time. | |
HIDE_TIME_TO_FIRST_RESPONSE |
False | If set to any value, the time to first response will not be displayed in the generated markdown file. | |
HIDE_TIME_TO_CLOSE |
False | If set to any value, the time to close will not be displayed in the generated markdown file. | |
HIDE_TIME_TO_ANSWER |
False | If set to any value, the time to answer a discussion will not be displayed in the generated markdown file. | |
HIDE_LABEL_METRICS |
False | If set to any value, the time in label metrics will not be displayed in the generated markdown file. | |
IGNORE_USERS |
False | A comma separated list of users to ignore when calculating metrics. (ie. IGNORE_USERS: 'user1,user2' ) |
This workflow searches for the issues created last month, and generates an issue with metrics.
name: Monthly issue metrics
on:
workflow_dispatch:
schedule:
- cron: '3 2 1 * *'
permissions:
issues: write
pull-requests: read
jobs:
build:
name: issue metrics
runs-on: ubuntu-latest
steps:
- name: Get dates for last month
shell: bash
run: |
# Calculate the first day of the previous month
first_day=$(date -d "last month" +%Y-%m-01)
# Calculate the last day of the previous month
last_day=$(date -d "$first_day +1 month -1 day" +%Y-%m-%d)
#Set an environment variable with the date range
echo "$first_day..$last_day"
echo "last_month=$first_day..$last_day" >> "$GITHUB_ENV"
- name: Run issue-metrics tool
uses: github/issue-metrics@v2
env:
GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
SEARCH_QUERY: 'repo:owner/repo is:issue created:${{ env.last_month }} -reason:"not planned"'
- name: Create issue
uses: peter-evans/create-issue-from-file@v4
with:
title: Monthly issue metrics report
token: ${{ secrets.GITHUB_TOKEN }}
content-filepath: ./issue_metrics.md
assignees: <YOUR_GITHUB_HANDLE_HERE>
This workflow searches for the issues created between 2023-05-01..2023-05-31, and generates an issue with metrics.
name: Monthly issue metrics
on:
workflow_dispatch:
permissions:
issues: write
pull-requests: read
jobs:
build:
name: issue metrics
runs-on: ubuntu-latest
steps:
- name: Run issue-metrics tool
uses: github/issue-metrics@v2
env:
GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
SEARCH_QUERY: 'repo:owner/repo is:issue created:2023-05-01..2023-05-31 -reason:"not planned"'
- name: Create issue
uses: peter-evans/create-issue-from-file@v4
with:
title: Monthly issue metrics report
token: ${{ secrets.GITHUB_TOKEN }}
content-filepath: ./issue_metrics.md
assignees: <YOUR_GITHUB_HANDLE_HERE>
This workflow searches for the issues created last month, and generates an issue with metrics. It also uses the GH_TOKEN
to scan a different repository than the one the workflow file is in.
name: Monthly issue metrics
on:
workflow_dispatch:
permissions:
issues: write
pull-requests: read
jobs:
build:
name: issue metrics
runs-on: ubuntu-latest
steps:
- name: Get dates for last month
shell: bash
run: |
# Calculate the first day of the previous month
first_day=$(date -d "last month" +%Y-%m-01)
# Calculate the last day of the previous month
last_day=$(date -d "$first_day +1 month -1 day" +%Y-%m-%d)
#Set an environment variable with the date range
echo "$first_day..$last_day"
echo "last_month=$first_day..$last_day" >> "$GITHUB_ENV"
- name: Get issue metrics (repo1)
uses: github/issue-metrics@v2
env:
GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
SEARCH_QUERY: 'repo:owner/repo is:issue created:2023-05-01..2023-05-31 -reason:"not planned"'
- name: Copy issue metrics (repo1)
run: |
cp ./issue_metrics.md ./issue_metrics_repo1.md
- name: Get issue metrics (repo2)
uses: github/issue-metrics@v2
env:
GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
SEARCH_QUERY: 'repo:owner/repo is:issue created:2023-05-01..2023-05-31 -reason:"not planned"'
- name: Copy issue metrics (repo2)
run: |
cp ./issue_metrics.md ./issue_metrics_repo2.md
- name: change owner of issue_metrics.md
run: |
sudo chown runner:runner ./issue_metrics.md
- name: Merge issue metrics
run: |
rm ./issue_metrics.md
echo "## repo1" >> ./issue_metrics.md
cat ./issue_metrics_repo1.md >> ./issue_metrics.md
echo "## repo2" >> ./issue_metrics.md
cat ./issue_metrics_repo2.md >> ./issue_metrics.md
sed -i '/# Issue Metrics/d' ./issue_metrics.md
sed -i '/_This report was generated with the [Issue Metrics Action](https://github.com/github/issue-metrics)_/d' ./issue_metrics.md
echo "_This report was generated with the [Issue Metrics Action](https://github.com/github/issue-metrics)_" >> ./issue_metrics.md
- name: Create issue
uses: peter-evans/create-issue-from-file@v4
with:
title: Monthly issue metrics report (dev)
token: ${{ secrets.GITHUB_TOKEN }}
content-filepath: ./issue_metrics.md
assignees: <YOUR_GITHUB_HANDLE_HERE>
Issues or Pull Requests? Open or closed? This action can be configured to run metrics on discussions, pull requests and/or issues. It is also configurable by whether they were open or closed in the specified time window. Further query options are listed in the documentation on searching issues and pull requests or the documentation on searching discussions. Search results are limited to 1000 results by the GitHub API. Here are some search query examples:
Issues opened in May 2023:
repo:owner/repo is:issue created:2023-05-01..2023-05-31
Issues closed in May 2023 (may have been open in May or earlier):
repo:owner/repo is:issue closed:2023-05-01..2023-05-31
Pull requests opened in May 2023:
repo:owner/repo is:pr created:2023-05-01..2023-05-31
Pull requests closed in May 2023 (may have been open in May or earlier):
repo:owner/repo is:pr closed:2023-05-01..2023-05-31
Discussions opened in May 2023:
repo:owner/repo type:discussions created:2023-05-01..2023-05-31
Discussions opened in May 2023 with category of engineering and label of question:
repo:owner/repo type:discussions created:2023-05-01..2023-05-31 category:engineering label:"question"
Both issues and pull requests opened in May 2023:
repo:owner/repo created:2023-05-01..2023-05-31
Both issues and pull requests closed in May 2023 (may have been open in May or earlier):
repo:owner/repo closed:2023-05-01..2023-05-31
OK, but what if I want both open or closed issues and pull requests? Due to limitations in issue search (no ability for OR logic), you will need to run the action twice, once for opened and once for closed. Here is an example workflow that does this:
name: Monthly issue metrics
on:
workflow_dispatch:
schedule:
- cron: '3 2 1 * *'
permissions:
issues: write
pull-requests: read
jobs:
build:
name: issue metrics
runs-on: ubuntu-latest
steps:
- name: Run issue-metrics tool for issues and prs opened in May 2023
uses: github/issue-metrics@v2
env:
GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
SEARCH_QUERY: 'repo:owner/repo created:2023-05-01..2023-05-31 -reason:"not planned"'
- name: Create issue for opened issues and prs
uses: peter-evans/create-issue-from-file@v4
with:
title: Monthly issue metrics report for opened issues and prs
token: ${{ secrets.GITHUB_TOKEN }}
content-filepath: ./issue_metrics.md
assignees: <YOUR_GITHUB_HANDLE_HERE>
- name: Run issue-metrics tool for issues and prs closed in May 2023
uses: github/issue-metrics@v2
env:
GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
SEARCH_QUERY: 'repo:owner/repo closed:2023-05-01..2023-05-31 -reason:"not planned"'
- name: Create issue for closed issues and prs
uses: peter-evans/create-issue-from-file@v4
with:
title: Monthly issue metrics report for closed issues and prs
content-filepath: ./issue_metrics.md
assignees: <YOUR_GITHUB_HANDLE_HERE>
Note: The discussions API currently doesn't support the LabeledEvent
so this action cannot measure the time spent in a label for discussions.
Sometimes it is helpful to know how long an issue or pull request spent in a particular label. This action can be configured to measure the time spent in a label. This is different from only wanting to measure issues with a specific label. If that is what you want, see the section on configuring your search query.
Here is an example workflow that does this:
name: Monthly issue metrics
on:
workflow_dispatch:
permissions:
issues: write
pull-requests: read
jobs:
build:
name: issue metrics
runs-on: ubuntu-latest
steps:
- name: Run issue-metrics tool
uses: github/issue-metrics@v2
env:
GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
LABELS_TO_MEASURE: 'waiting-for-manager-approval,waiting-for-security-review'
SEARCH_QUERY: 'repo:owner/repo is:issue created:2023-05-01..2023-05-31 -reason:"not planned"'
- name: Create issue
uses: peter-evans/create-issue-from-file@v4
with:
title: Monthly issue metrics report
token: ${{ secrets.GITHUB_TOKEN }}
content-filepath: ./issue_metrics.md
assignees: <YOUR_GITHUB_HANDLE_HERE>
then the report will look like this:
# Issue Metrics
| Metric | Value |
| --- | ---: |
| Average time to first response | 0:50:44.666667 |
| Average time to close | 6 days, 7:08:52 |
| Average time to answer | 1 day |
| Average time spent in waiting-for-manager-approval | 0:00:41 |
| Average time spent in waiting-for-security-review | 2 days, 4:25:03 |
| Number of items that remain open | 2 |
| Number of items closed | 1 |
| Total number of items created | 3 |
| Title | URL | Time to first response | Time to close | Time to answer | Time spent in waiting-for-manager-approval | Time spent in waiting-for-security-review |
| --- | --- | --- | --- | --- | --- | --- |
| Pull Request Title 1 | https://github.com/user/repo/pulls/1 | 0:05:26 | None | None | None | None |
| Issue Title 2 | https://github.com/user/repo/issues/2 | 2:26:07 | None | None | 0:00:41 | 2 days, 4:25:03 |
Here is the output with no hidden columns:
# Issue Metrics
| Metric | Value |
| --- | ---: |
| Average time to first response | 0:50:44.666667 |
| Average time to close | 6 days, 7:08:52 |
| Average time to answer | 1 day |
| Number of items that remain open | 2 |
| Number of items closed | 1 |
| Total number of items created | 3 |
| Title | URL | Time to first response | Time to close | Time to answer |
| --- | --- | --- | --- | --- |
| Discussion Title 1 | https://github.com/user/repo/discussions/1 | 0:00:41 | 6 days, 7:08:52 | 1 day |
| Pull Request Title 2 | https://github.com/user/repo/pulls/2 | 0:05:26 | None | None |
| Issue Title 3 | https://github.com/user/repo/issues/3 | 2:26:07 | None | None |
Here is the output with all hidable columns hidden:
# Issue Metrics
| Metric | Value |
| --- | ---: |
| Number of items that remain open | 2 |
| Number of items closed | 1 |
| Total number of items created | 3 |
| Title | URL |
| --- | --- |
| Discussion Title 1 | https://github.com/user/repo/discussions/1 |
| Pull Request Title 2 | https://github.com/user/repo/pulls/2 |
| Issue Title 3 | https://github.com/user/repo/issues/3 |
There is JSON output available as well. You could use it for any number of possibilities, but here is one example that demonstrates retreiving the JSON output and then printing it out.
name: Monthly issue metrics
on:
workflow_dispatch:
schedule:
- cron: '3 2 1 * *'
permissions:
issues: write
pull-requests: read
jobs:
build:
name: issue metrics
runs-on: ubuntu-latest
steps:
- name: Run issue-metrics tool
id: issue-metrics
uses: github/issue-metrics@v2
env:
GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
SEARCH_QUERY: 'repo:owner/repo is:issue created:2023-05-01..2023-05-31 -reason:"not planned"'
- name: Print output of issue metrics tool
run: echo "${{ steps.issue-metrics.outputs.metrics }}"
The assignee part of this workflow action comes from a different GitHub action and currently GitHub issues don't support assigning groups.
By way of work around, you could use the GitHub API to retrieve the members of the team and then put them in a comma separated string that you provide as the assignee. This requires setting up a new GitHub API token (referred to below as CUSTOM_TOKEN
) which has read:org
permissions assigned and single sign on authorization as needed. To do this, create a GitHub API token with permissions to read the org (read:org
). Then take the value of the API token you just created, and create a repository secret where the name of the secret is CUSTOM_TOKEN
and the value of the secret the API token.
That might look something like the workflow below where ORG
is your organization name and TEAM_SLUG
is the name of the team:
name: Monthly issue metrics
on:
workflow_dispatch:
schedule:
- cron: '3 2 1 * *'
permissions:
issues: write
pull-requests: read
jobs:
build:
name: issue metrics
runs-on: ubuntu-latest
steps:
- name: Get dates for last month
shell: bash
run: |
# Calculate the first day of the previous month
first_day=$(date -d "last month" +%Y-%m-01)
# Calculate the last day of the previous month
last_day=$(date -d "$first_day +1 month -1 day" +%Y-%m-%d)
#Set an environment variable with the date range
echo "$first_day..$last_day"
echo "last_month=$first_day..$last_day" >> "$GITHUB_ENV"
- name: Run issue-metrics tool
uses: github/issue-metrics@v2
env:
GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
SEARCH_QUERY: 'repo:owner/repo is:issue created:${{ env.last_month }} -reason:"not planned"'
- name: Get user names from team
run: |
teamMembers="$(gh api /orgs/ORG/teams/TEAM_SLUG/members | jq -r '.[].login' | paste -sd, -)"
echo 'TEAM_MEMBERS='$teamMembers >> $GITHUB_ENV
env:
GITHUB_TOKEN: ${{ secrets.CUSTOM_TOKEN }}
- name: Create issue
uses: peter-evans/create-issue-from-file@v4
with:
title: Monthly issue metrics report
token: ${{ secrets.GITHUB_TOKEN }}
content-filepath: ./issue_metrics.md
assignees: ${{ env.TEAM_MEMBERS }}
- Copy
.env-example
to.env
- Fill out the
.env
file with a token from a user that has access to the organization to scan (listed below). Tokens should have admin:org or read:org access. - Fill out the
.env
file with the search_query to filter issues by pip install -r requirements.txt
- Run
python3 ./issue_metrics.py
, which will output issue metrics data