cache Documentation

File: README.md Copy Chapter

Cache action

This action allows caching dependencies and build outputs to improve workflow execution time.

Two other actions are available in addition to the primary cache action:

• Restore action
• Save action

Documentation Copy Chapter

See "Caching dependencies to speed up workflows".

What's New Copy Chapter

⚠️ Important changes Copy Section

[!IMPORTANT]
actions/cache@v5 runs on the Node.js 24 runtime and requires a minimum Actions Runner version of 2.327.1.
If you are using self-hosted runners, ensure they are updated before upgrading.

The cache backend service has been rewritten from the ground up for improved performance and reliability. actions/cache now integrates with the new cache service (v2) APIs.

The new service will gradually roll out as of February 1st, 2025. The legacy service will also be sunset on the same date. Changes in these releases are fully backward compatible.

We are deprecating some versions of this action. We recommend upgrading to version v4 or v3 as soon as possible before February 1st, 2025. (Upgrade instructions below).

If you are using pinned SHAs, please use the SHAs of versions v4.2.0 or v3.4.0.

If you do not upgrade, all workflow runs using any of the deprecated actions/cache will fail.

Upgrading to the recommended versions will not break your workflows.

Additionally, if you are managing your own GitHub runners, you must update your runner version to 2.231.0 or newer to ensure compatibility with the new cache service.
Failure to update both the action version and your runner version may result in workflow failures after the migration date.

Read more about the change & access the migration guide: reference to the announcement.

v6 Copy Section

• Updated @actions/cache, @actions/core, @actions/exec to latest major versions
• Migrated to ESM module system

v5 Copy Section

• Updated to node 24
• Requires a minimum Actions Runner version of 2.327.1

v4 Copy Section

• Integrated with the new cache service (v2) APIs.
• Updated to node 20

v3 Copy Section

• Integrated with the new cache service (v2) APIs.
• Added support for caching in GHES 3.5+.
• Fixed download issue for files > 2GB during restore.
• Updated the minimum runner version support from node 12 -> node 16.
• Fixed avoiding empty cache save when no files are available for caching.
• Fixed tar creation error while trying to create tar with path as ~/ home folder on ubuntu-latest.
• Fixed zstd failing on amazon linux 2.0 runners.
• Fixed cache not working with github workspace directory or current directory.
• Fixed the download stuck problem by introducing a timeout of 1 hour for cache downloads.
• Fix zstd not working for windows on gnu tar in issues.
• Allowing users to provide a custom timeout as input for aborting download of a cache segment using an environment variable SEGMENT_DOWNLOAD_TIMEOUT_MINS. Default is 10 minutes.
• New actions are available for granular control over caches - restore and save.
• Support cross-os caching as an opt-in feature. See Cross OS caching for more info.
• Added option to fail job on cache miss. See Exit workflow on cache miss for more info.
• Fix zstd not being used after zstd version upgrade to 1.5.4 on hosted runners
• Added option to lookup cache without downloading it.
• Reduced segment size to 128MB and segment timeout to 10 minutes to fail fast in case the cache download is stuck.

See the v2 README.md for older updates.

Usage Copy Chapter

Pre-requisites Copy Section

Create a workflow .yml file in your repository's .github/workflows directory. An example workflow is available below. For more information, see the GitHub Help Documentation for Creating a workflow file.

If you are using this inside a container, a POSIX-compliant tar needs to be included and accessible from the execution path.

Note: actions/cache@v5 runs on Node.js 24 and requires a minimum Actions Runner version of 2.327.1.

If you are using a self-hosted Windows runner, GNU tar and zstd are required for Cross-OS caching to work. They are also recommended to be installed in general so the performance is on par with hosted Windows runners.

Inputs Copy Section

• key - An explicit key for a cache entry. See creating a cache key.
• path - A list of files, directories, and wildcard patterns to cache and restore. See @actions/glob for supported patterns.
• restore-keys - An ordered multiline string listing the prefix-matched keys, that are used for restoring stale cache if no cache hit occurred for key.
• enableCrossOsArchive - An optional boolean when enabled, allows Windows runners to save or restore caches that can be restored or saved respectively on other platforms. Default: false
• fail-on-cache-miss - Fail the workflow if cache entry is not found. Default: false
• lookup-only - If true, only checks if cache entry exists and skips download. Does not change save cache behavior. Default: false

Environment Variables

• SEGMENT_DOWNLOAD_TIMEOUT_MINS - Segment download timeout (in minutes, default 10) to abort download of the segment if not completed in the defined number of minutes. Read more

Outputs Copy Section

• cache-hit - A string value to indicate an exact match was found for the key.
  • If there's a cache hit, this will be 'true' or 'false' to indicate if there's an exact match for key.
  • If there's a cache miss, this will be an empty string.

See Skipping steps based on cache-hit for info on using this output

Cache scopes Copy Section

The cache is scoped to the key, version, and branch. The default branch cache is available to other branches.

See Matching a cache key for more info.

Read-only access Copy Section

Some workflow runs only have read-only access to the cache. A common case is a workflow triggered by a pull request from a fork: such runs can restore existing caches but may not be permitted to save new ones.

When the cache token is read-only, the save step does not fail the job. Instead, @actions/cache reports the denial once as a warning (for example, Failed to save: ... cache write denied: ...) and the step completes successfully without writing a cache entry. Restores in the same run continue to work as usual.

Note This applies to the action's normal save path as well as the standalone Save action. If you intentionally want a restore-only setup, see Make cache read only / Reuse cache from centralized job.

Example cache workflow Copy Section

Restoring and saving cache using a single action

name: Caching Primes

on: push

jobs:
  build:
    runs-on: ubuntu-latest

    steps:
    - uses: actions/checkout@v6

    - name: Cache Primes
      id: cache-primes
      uses: actions/cache@v6
      with:
        path: prime-numbers
        key: ${{ runner.os }}-primes

    - name: Generate Prime Numbers
      if: steps.cache-primes.outputs.cache-hit != 'true'
      run: /generate-primes.sh -d prime-numbers

    - name: Use Prime Numbers
      run: /primes.sh -d prime-numbers

The cache action provides a cache-hit output which is set to true when the cache is restored using the primary key and false when the cache is restored using restore-keys or no cache is restored.

Using a combination of restore and save actions

name: Caching Primes

on: push

jobs:
  build:
    runs-on: ubuntu-latest

    steps:
    - uses: actions/checkout@v6

    - name: Restore cached Primes
      id: cache-primes-restore
      uses: actions/cache/restore@v6
      with:
        path: |
          path/to/dependencies
          some/other/dependencies
        key: ${{ runner.os }}-primes
    .
    . //intermediate workflow steps
    .
    - name: Save Primes
      id: cache-primes-save
      uses: actions/cache/save@v6
      with:
        path: |
          path/to/dependencies
          some/other/dependencies
        key: ${{ steps.cache-primes-restore.outputs.cache-primary-key }}

Note
You must use the cache or restore action in your workflow before you need to use the files that might be restored from the cache. If the provided key matches an existing cache, a new cache is not created and if the provided key doesn't match an existing cache, a new cache is automatically created provided the job completes successfully.

Caching Strategies Copy Chapter

With the introduction of the restore and save actions, a lot of caching use cases can now be achieved. Please see the caching strategies document for understanding how you can use the actions strategically to achieve the desired goal.

Implementation Examples Copy Chapter

Every programming language and framework has its own way of caching.

See Examples for a list of actions/cache implementations for use with:

• Bun
• C# - NuGet
• Clojure - Lein Deps
• D - DUB
• Deno
• Elixir - Mix
• Go - Modules
• Haskell - Cabal
• Haskell - Stack
• Java - Gradle
• Java - Maven
• Node - npm
• Node - Lerna
• Node - Yarn
• OCaml/Reason - esy
• PHP - Composer
• Python - pip
• Python - pipenv
• R - renv
• Ruby - Bundler
• Rust - Cargo
• Scala - SBT
• Swift, Objective-C - Carthage
• Swift, Objective-C - CocoaPods
• Swift - Swift Package Manager
• Swift - Mint

Creating a cache key Copy Chapter

A cache key can include any of the contexts, functions, literals, and operators supported by GitHub Actions.

For example, using the hashFiles function allows you to create a new cache when dependencies change.

- uses: actions/cache@v6
    with:
      path: |
        path/to/dependencies
        some/other/dependencies
      key: ${{ runner.os }}-${{ hashFiles('**/lockfiles') }}

Additionally, you can use arbitrary command output in a cache key, such as a date or software version:

# http://man7.org/linux/man-pages/man1/date.1.html
  - name: Get Date
    id: get-date
    run: |
      echo "date=$(/bin/date -u "+%Y%m%d")" >> $GITHUB_OUTPUT
    shell: bash

  - uses: actions/cache@v6
    with:
      path: path/to/dependencies
      key: ${{ runner.os }}-${{ steps.get-date.outputs.date }}-${{ hashFiles('**/lockfiles') }}

See Using contexts to create cache keys

Cache Limits Copy Chapter

A repository can have up to 10GB of caches. Once the 10GB limit is reached, older caches will be evicted based on when the cache was last accessed. Caches that are not accessed within the last week will also be evicted.

Skipping steps based on cache-hit Copy Chapter

Using the cache-hit output, subsequent steps (such as install or build) can be skipped when a cache hit occurs on the key. It is recommended to install missing/updated dependencies in case of a partial key match when the key is dependent on the hash of the package file.

Example:

steps:
  - uses: actions/checkout@v6

  - uses: actions/cache@v6
    id: cache
    with:
      path: path/to/dependencies
      key: ${{ runner.os }}-${{ hashFiles('**/lockfiles') }}

  - name: Install Dependencies
    if: steps.cache.outputs.cache-hit != 'true'
    run: /install.sh

Note The id defined in actions/cache must match the id in the if statement (i.e. steps.[ID].outputs.cache-hit)

Cache Version Copy Chapter

Cache version is a hash generated for a combination of compression tool used (Gzip, Zstd, etc. based on the runner OS) and the path of directories being cached. If two caches have different versions, they are identified as unique caches while matching. This, for example, means that a cache created on a windows-latest runner can't be restored on ubuntu-latest as cache Versions are different.

Pro tip: The list caches API can be used to get the version of a cache. This can be helpful to troubleshoot cache miss due to version.

Example
The workflow will create 3 unique caches with same keys. Ubuntu and windows runners will use different compression technique and hence create two different caches. And build-linux will create two different caches as the paths are different.

jobs:
  build-linux:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v6

      - name: Cache Primes
        id: cache-primes
        uses: actions/cache@v6
        with:
          path: prime-numbers
          key: primes

      - name: Generate Prime Numbers
        if: steps.cache-primes.outputs.cache-hit != 'true'
        run: ./generate-primes.sh -d prime-numbers

      - name: Cache Numbers
        id: cache-numbers
        uses: actions/cache@v6
        with:
          path: numbers
          key: primes

      - name: Generate Numbers
        if: steps.cache-numbers.outputs.cache-hit != 'true'
        run: ./generate-primes.sh -d numbers

  build-windows:
    runs-on: windows-latest
    steps:
      - uses: actions/checkout@v6

      - name: Cache Primes
        id: cache-primes
        uses: actions/cache@v6
        with:
          path: prime-numbers
          key: primes

      - name: Generate Prime Numbers
        if: steps.cache-primes.outputs.cache-hit != 'true'
        run: ./generate-primes -d prime-numbers

Known practices and workarounds Copy Chapter

There are a number of community practices/workarounds to fulfill specific requirements. You may choose to use them if they suit your use case. Note these are not necessarily the only solution or even a recommended solution.

• Cache segment restore timeout
• Update a cache
• Use cache across feature branches
• Cross OS cache
• Force deletion of caches overriding default cache eviction policy

Windows environment variables Copy Section

Please note that Windows environment variables (like %LocalAppData%) will NOT be expanded by this action. Instead, prefer using ~ in your paths which will expand to the HOME directory. For example, instead of %LocalAppData%, use ~\AppData\Local. For a list of supported default environment variables, see the Learn GitHub Actions: Variables page.

Note Copy Chapter

Thank you for your interest in this GitHub repo, however, right now we are not taking contributions.

We continue to focus our resources on strategic areas that help our customers be successful while making developers' lives easier. While GitHub Actions remains a key part of this vision, we are allocating resources towards other areas of Actions and are not taking contributions to this repository at this time. The GitHub public roadmap is the best place to follow along for any updates on features we’re working on and what stage they’re in.

We are taking the following steps to better direct requests related to GitHub Actions, including:

1. We will be directing questions and support requests to our Community Discussions area

2. High Priority bugs can be reported through Community Discussions or you can report these to our support team https://support.github.com/contact/bug-report.

3. Security Issues should be handled as per our security.md.

We will still provide security updates for this project and fix major breaking changes during this time.

You are welcome to still raise bugs in this repo.

License Copy Chapter

The scripts and documentation in this project are released under the MIT License

File: restore/README.md Copy Chapter

Restore action

The restore action restores a cache. It works similarly to the cache action except that it doesn't have a post step to save the cache. This action provides granular ability to restore a cache without having to save it. It accepts the same set of inputs as the cache action.

Documentation Copy Chapter

Inputs Copy Section

• key - An explicit key for a cache entry. See creating a cache key.
• path - A list of files, directories, and wildcard patterns to restore. See @actions/glob for supported patterns.
• restore-keys - An ordered list of prefix-matched keys to use for restoring stale cache if no cache hit occurred for key.
• fail-on-cache-miss - Fail the workflow if cache entry is not found. Default: false
• lookup-only - If true, only checks if cache entry exists and skips download. Default: false

Outputs Copy Section

• cache-hit - A boolean value to indicate an exact match was found for the key.
• cache-primary-key - Cache primary key passed in the input to use in subsequent steps of the workflow.
• cache-matched-key - Key of the cache that was restored, it could either be the primary key on cache-hit or a partial/complete match of one of the restore keys.

Note
cache-hit will be set to true only when cache hit occurs for the exact key match. For a partial key match via restore-keys or a cache miss, it will be set to false.

Environment Variables Copy Section

• SEGMENT_DOWNLOAD_TIMEOUT_MINS - Segment download timeout (in minutes, default 10) to abort download of the segment if not completed in the defined number of minutes. Read more

Use cases Copy Chapter

As this is a newly introduced action to give users more control in their workflows, below are some use cases where one can use this action.

Only restore cache Copy Section

If you are using separate jobs to create and save your cache(s) to be reused by other jobs in a repository, this action will take care of your cache restoring needs.

steps:
  - uses: actions/checkout@v6

  - uses: actions/cache/restore@v5
    id: cache
    with:
      path: path/to/dependencies
      key: ${{ runner.os }}-${{ hashFiles('**/lockfiles') }}

  - name: Install Dependencies
    if: steps.cache.outputs.cache-hit != 'true'
    run: /install.sh

  - name: Build
    run: /build.sh

  - name: Publish package to public
    run: /publish.sh

Once the cache is restored, unlike actions/cache, this action won't run a post step to do post-processing, and the rest of the workflow will run as usual.

Save intermediate private build artifacts Copy Section

In case of multi-module projects, where the built artifact of one project needs to be reused in subsequent child modules, the need to rebuild the parent module again and again with every build can be eliminated. The actions/cache or actions/cache/save action can be used to build and save the parent module artifact once, and it can be restored multiple times while building the child modules.

Step 1 - Build the parent module and save it

steps:
  - uses: actions/checkout@v6

  - name: Build
    run: /build-parent-module.sh

  - uses: actions/cache/save@v5
    id: cache
    with:
      path: path/to/dependencies
      key: ${{ runner.os }}-${{ hashFiles('**/lockfiles') }}

Step 2 - Restore the built artifact from cache using the same key and path

steps:
  - uses: actions/checkout@v6

  - uses: actions/cache/restore@v5
    id: cache
    with:
      path: path/to/dependencies
      key: ${{ runner.os }}-${{ hashFiles('**/lockfiles') }}

  - name: Install Dependencies
    if: steps.cache.outputs.cache-hit != 'true'
    run: /install.sh

  - name: Build
    run: /build-child-module.sh

  - name: Publish package to public
    run: /publish.sh

Exit workflow on cache miss Copy Section

You can use fail-on-cache-miss: true to exit a workflow on a cache miss. This way you can restrict your workflow to only build when there is a cache-hit.

To fail if there is no cache hit for the primary key, leave restore-keys empty!

steps:
  - uses: actions/checkout@v6

  - uses: actions/cache/restore@v5
    id: cache
    with:
      path: path/to/dependencies
      key: ${{ runner.os }}-${{ hashFiles('**/lockfiles') }}
      fail-on-cache-miss: true

  - name: Build
    run: /build.sh

Tips Copy Chapter

Reusing primary key and restored key in the save action Copy Section

Usually you may want to use the same key with both actions/cache/restore and actions/cache/save actions. To achieve this, use outputs from the restore action to reuse the same primary key (or the key of the cache that was restored).

Using restore action outputs to make save action behave just like the cache action Copy Section

The outputs cache-primary-key and cache-matched-key can be used to check if the restored cache is same as the given primary key. Alternatively, the cache-hit output can also be used to check if the restored was a complete match or a partially restored cache.

Ensuring proper restores and save happen across the actions Copy Section

It is very important to use the same key and path that were used by either actions/cache or actions/cache/save while saving the cache. Learn more about cache key naming and versioning here.

File: save/README.md Copy Chapter

Save action

The save action saves a cache. It works similarly to the cache action except that it doesn't first do a restore. This action provides granular ability to save a cache without having to restore it, or to do a save at any stage of the workflow job -- not only in post phase.

Documentation Copy Chapter

Inputs Copy Section

• key - An explicit key for a cache entry. See creating a cache key.
• path - A list of files, directories, and wildcard patterns to cache. See @actions/glob for supported patterns.
• upload-chunk-size - The chunk size used to split up large files during upload, in bytes

Outputs Copy Section

This action has no outputs.

Use cases Copy Chapter

Only save cache Copy Section

If you are using separate jobs for generating common artifacts and sharing them across jobs, this action will take care of your cache saving needs.

steps:
  - uses: actions/checkout@v6

  - name: Install Dependencies
    run: /install.sh

  - name: Build artifacts
    run: /build.sh

  - uses: actions/cache/save@v5
    id: cache
    with:
      path: path/to/dependencies
      key: ${{ runner.os }}-${{ hashFiles('**/lockfiles') }}

Re-evaluate cache key while saving Copy Section

With this save action, the key can now be re-evaluated while executing the action. This helps in cases where lockfiles are generated during the build.

Let's say we have a restore step that computes a key at runtime.

Restore a cache

uses: actions/cache/restore@v5
id: restore-cache
with:
    key: cache-${{ hashFiles('**/lockfiles') }}

Case 1 - Where a user would want to reuse the key as it is

uses: actions/cache/save@v5
with:
    key: ${{ steps.restore-cache.outputs.cache-primary-key }}

Case 2 - Where the user would want to re-evaluate the key

uses: actions/cache/save@v5
with:
    key: npm-cache-${{hashfiles(package-lock.json)}}

Always save cache Copy Section

There are instances where some flaky test cases would fail the entire workflow and users would get frustrated because the builds would run for hours and the cache couldn't be saved as the workflow failed in between.
For such use-cases, users now have the ability to use the actions/cache/save action to save the cache by using an always() condition.
This way the cache will always be saved if generated, or a warning will be generated that nothing is found on the cache path. Users can also use the if condition to only execute the actions/cache/save action depending on the output of previous steps. This way they get more control of when to save the cache.

To avoid saving a cache that already exists, the cache-hit output from a restore step should be checked.

The cache-primary-key output from the restore step should also be used to ensure
the cache key does not change during the build if it's calculated based on file contents.

Here's an example where we imagine we're calculating a lot of prime numbers and want to cache them:

name: Always Caching Prime Numbers

on: push

jobs:
  build:
    runs-on: ubuntu-latest

    steps:
    - uses: actions/checkout@v6

    - name: Restore cached Prime Numbers
      id: cache-prime-numbers-restore
      uses: actions/cache/restore@v5
      with:
        key: ${{ runner.os }}-prime-numbers
        path: |
          path/to/dependencies
          some/other/dependencies

    # Intermediate workflow steps

    - name: Always Save Prime Numbers
      id: cache-prime-numbers-save
      if: always() && steps.cache-prime-numbers-restore.outputs.cache-hit != 'true'
      uses: actions/cache/save@v5
      with:
        key: ${{ steps.cache-prime-numbers-restore.outputs.cache-primary-key }}
        path: |
          path/to/dependencies
          some/other/dependencies