feat(backend): implement comprehensive caching layer for all GET endpoints (Part 2) #10979

Swiftyos · 2025-09-24T14:24:54Z

Summary

Extends the caching implementation from Part 1 to cover all GET endpoints across the platform
Separates cache logic into dedicated cache.py files for better code organization and maintainability
Implements intelligent cache invalidation on all mutation operations (CREATE/UPDATE/DELETE)

Changes

New Cache Modules

backend/server/routers/cache.py - V1 API endpoint caching
backend/server/v2/library/cache.py - Library API caching
backend/server/v2/store/cache.py - Store API caching (refactored from routes)

Cached Endpoints

Graphs: List and details (15-30 min TTL)
Executions: Graph executions (5 min TTL)
User Data: Preferences and timezone (30-60 min TTL)
Library: Agents, favorites, and presets (10-30 min TTL)
Store: Listings and profiles (5-60 min TTL)

Key Features

Selective caching only for default queries (bypasses cache for filtered/searched results)
Proper cache invalidation using positional arguments to match function calls
Configurable TTLs based on data volatility
Built-in thundering herd protection via the @cached decorator

Test Plan

All 20 cache-specific tests passing
Cache hit/miss behavior validated
Cache invalidation on mutations verified
Integration tests confirm proper operation
Manual testing of affected endpoints

Performance Impact

Reduces database load for frequently accessed data
Improves response times for cached endpoints
Maintains data consistency through intelligent invalidation

…oints (Part 2) - Created separate cache.py modules for better code organization - backend/server/routers/cache.py for V1 API endpoints - backend/server/v2/library/cache.py for library endpoints - backend/server/v2/store/cache.py (refactored from routes) - Added caching to all major GET endpoints: - Graphs list/details with 15-30 min TTL - Graph executions with 5 min TTL - User preferences/timezone with 30-60 min TTL - Library agents/favorites/presets with 10-30 min TTL - Store listings/profiles with 5-60 min TTL - Implemented intelligent cache invalidation: - Clears relevant caches on CREATE/UPDATE/DELETE operations - Uses positional arguments for cache_delete to match function calls - Selective caching only for default queries (bypasses cache for filtered/searched results) - Added comprehensive test coverage: - 20 cache-specific tests all passing - Validates cache hit/miss behavior - Verifies invalidation on mutations - Performance improvements: - Reduces database load for frequently accessed data - Built-in thundering herd protection via @cached decorator - Configurable TTLs based on data volatility

netlify · 2025-09-24T14:25:04Z

✅ Deploy Preview for auto-gpt-docs-dev canceled.

Name	Link
🔨 Latest commit	`045009a`
🔍 Latest deploy log	https://app.netlify.com/projects/auto-gpt-docs-dev/deploys/68e017b1a55e1b0008183f2f

netlify · 2025-09-24T14:25:10Z

✅ Deploy Preview for auto-gpt-docs canceled.

Name	Link
🔨 Latest commit	`045009a`
🔍 Latest deploy log	https://app.netlify.com/projects/auto-gpt-docs/deploys/68e017b1aa078d00087b48de

qodo-merge-pro · 2025-09-24T14:25:32Z

PR Reviewer Guide 🔍

Here are some key observations to aid the review process:

⏱️ Estimated effort to review: 3 🔵🔵🔵⚪⚪
🧪 PR contains tests
🔒 No security concerns identified
⚡ Recommended focus areas for review Behavior Change The list graphs endpoint previously passed a 'filter_by="active"' argument to the DB call; the new cached path omits this filter. Confirm parity with previous behavior or reintroduce the filter to avoid returning inactive graphs. async def list_graphs( user_id: Annotated[str, Security(get_user_id)], ) -> Sequence[graph_db.GraphMeta]: paginated_result = await cache.get_cached_graphs( user_id=user_id, page=1, page_size=250, ) return paginated_result.graphs Export Path Bypass The cached get_graph path does not pass include_subgraphs/for_export flags. Ensure non-export requests do not require subgraphs; otherwise, some consumers may get incomplete graph data compared to the previous always-include_subgraphs implementation. async def get_graph( graph_id: str, user_id: Annotated[str, Security(get_user_id)], version: int \| None = None, for_export: bool = False, ) -> graph_db.GraphModel: # Use cache for non-export requests if not for_export: graph = await cache.get_cached_graph( graph_id=graph_id, version=version, user_id=user_id, ) else: graph = await graph_db.get_graph( graph_id, version, user_id=user_id, for_export=for_export, include_subgraphs=True, # needed to construct full credentials input schema ) if not graph: raise HTTPException(status_code=404, detail=f"Graph #{graph_id} not found.") return graph Cache Selectivity Caching for list_library_agents only applies when sort_by is UPDATED_AT and no search term. Consider also excluding non-default page_size or page to ensure TTL assumptions; validate that ordering defaults are identical to DB defaults to prevent subtle mismatches. Returns: A LibraryAgentResponse containing agents and pagination metadata. Raises: HTTPException: If a server/database error occurs. """ try: # Use cache for default queries (no search term, default sort) if search_term is None and sort_by == library_model.LibraryAgentSort.UPDATED_AT: return await library_cache.get_cached_library_agents( user_id=user_id, page=page, page_size=page_size, ) else: # Direct DB query for searches and custom sorts return await library_db.list_library_agents( user_id=user_id, search_term=search_term, sort_by=sort_by, page=page, page_size=page_size, ) except Exception as e:

deepsource-io · 2025-09-24T14:25:39Z

Here's the code health analysis summary for commits 8b4eb6f..045009a. View details on DeepSource ↗.

Analysis Summary

Analyzer	Status	Summary	Link
JavaScript	✅ Success	❗ 1 occurence introduced	View Check ↗
Python	✅ Success	❗ 41 occurences introduced 🎯 1 occurence resolved	View Check ↗

💡 If you’re a repository administrator, you can configure the quality gates from the settings.

…stead of db The test was failing because routes now use cached functions. Updated the mock to patch the cache function which is what the route actually calls.

Copilot

Pull Request Overview

This PR implements comprehensive caching for all GET endpoints across the AutoGPT platform by separating cache logic into dedicated modules for better maintainability. The changes extend the caching implementation to cover graphs, executions, user data, library operations, and store listings with configurable TTLs based on data volatility.

Key changes:

Extracts cache functions into separate modules (cache.py files) from route handlers
Implements intelligent cache invalidation on all CREATE/UPDATE/DELETE operations
Adds comprehensive test coverage for cache behavior and invalidation patterns

Reviewed Changes

Copilot reviewed 10 out of 10 changed files in this pull request and generated 7 comments.

Show a summary per file

File	Description
`autogpt_platform/backend/backend/server/v2/store/routes.py`	Removes cache functions and imports them from new cache module
`autogpt_platform/backend/backend/server/v2/store/cache.py`	New cache module for Store API with all caching functions
`autogpt_platform/backend/backend/server/v2/library/routes_test.py`	Updates test to mock cache functions instead of DB functions
`autogpt_platform/backend/backend/server/v2/library/routes/presets.py`	Integrates caching for preset operations with cache invalidation
`autogpt_platform/backend/backend/server/v2/library/routes/agents.py`	Integrates caching for agent operations with cache invalidation
`autogpt_platform/backend/backend/server/v2/library/cache_test.py`	New comprehensive test suite for library cache invalidation
`autogpt_platform/backend/backend/server/v2/library/cache.py`	New cache module for Library API with all caching functions
`autogpt_platform/backend/backend/server/routers/v1.py`	Integrates caching for V1 API endpoints with cache invalidation
`autogpt_platform/backend/backend/server/routers/cache_test.py`	New comprehensive test suite for V1 API cache invalidation
`autogpt_platform/backend/backend/server/routers/cache.py`	New cache module for V1 API with all caching functions

autogpt_platform/backend/backend/server/routers/v1.py

autogpt_platform/backend/backend/server/v2/library/routes/presets.py

autogpt_platform/backend/backend/server/v2/library/routes/agents.py

The cached graph function was missing include_subgraphs=True parameter which is needed to construct full credentials input schema. This was causing test_access_store_listing_graph to fail.

… permissions When a graph is not found/accessible, we now clear the cache entry rather than caching the None result. This prevents issues with store listing permissions where a graph becomes accessible after approval but the cache still returns the old 'not found' result.

Swiftyos · 2025-09-26T09:21:10Z

unfavouriting a library agent is not working

github-actions · 2025-09-29T05:46:21Z

This pull request has conflicts with the base branch, please resolve those so we can evaluate the pull request.

coderabbitai · 2025-10-01T12:53:11Z

Important

Review skipped

Auto reviews are disabled on this repository.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Note

Other AI code review bot(s) detected

CodeRabbit has detected other AI code review bot(s) in this pull request and will avoid duplicating their findings in the review comments. This may lead to a less comprehensive review.

✨ Finishing touches

🧪 Generate unit tests (beta)

Create PR with unit tests
Post copyable unit tests in a comment
Commit unit tests in branch swiftyos/caching-pt2

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

…s/AutoGPT into swiftyos/caching-pt2

majdyz · 2025-10-02T01:48:48Z

autogpt_platform/backend/backend/server/v2/library/routes/presets.py

    merged_node_input = preset.inputs | inputs
    merged_credential_inputs = preset.credentials | credential_inputs

+    for page in range(1, 10):


I don't. get when you remove 10 pages and remove 5 pages for others, what's the factor?

majdyz · 2025-10-02T01:49:33Z

autogpt_platform/backend/backend/server/v2/library/routes/presets.py

+            result = await db.create_preset_from_graph_execution(user_id, preset)
+
+        # Clear presets list cache after creating new preset
+        for page in range(1, 5):


at this point, you can make a helper out of this :D to remove pages.

majdyz · 2025-10-02T01:51:14Z

autogpt_platform/backend/backend/server/v2/library/routes/agents.py

+        library_cache.get_cached_library_agent.cache_delete(
+            library_agent_id=library_agent_id, user_id=user_id
+        )
+        for page in range(1, 20):


here it's 20 :o

…ng more robust

Swiftyos requested a review from a team as a code owner September 24, 2025 14:24

Swiftyos requested review from Pwuts and kcze and removed request for a team September 24, 2025 14:24

github-project-automation bot added this to AutoGPT development kanban Sep 24, 2025

github-project-automation bot moved this to 🆕 Needs initial review in AutoGPT development kanban Sep 24, 2025

github-actions bot added the platform/backend AutoGPT Platform - Back end label Sep 24, 2025

github-actions bot added the size/xl label Sep 24, 2025

qodo-merge-pro bot added the Review effort 3/5 label Sep 24, 2025

fix(backend): update library favorites test to mock cache function in…

9147c2d

…stead of db The test was failing because routes now use cached functions. Updated the mock to patch the cache function which is what the route actually calls.

Swiftyos requested a review from Copilot September 24, 2025 14:40

Copilot AI reviewed Sep 24, 2025

View reviewed changes

Swiftyos added 11 commits September 24, 2025 16:51

fix(backend): include subgraphs in cached graph retrieval

ee4feff

The cached graph function was missing include_subgraphs=True parameter which is needed to construct full credentials input schema. This was causing test_access_store_listing_graph to fail.

Merge remote-tracking branch 'origin/dev' into swiftyos/caching-pt2

c61af53

fixed caching when adding agent from library

863e213

fixed caching when saving an agent

d68a3a1

fix duplicated caching code

12382e7

fmt

872ef5f

updated caching invalidation rules

bc5eb8a

Merge branch 'dev' into swiftyos/caching-pt2

d124c93

fix test

1afebcf

Merge branch 'dev' into swiftyos/caching-pt2

be1947f

Swiftyos added 2 commits September 29, 2025 09:55

fix merge error

30cdf9f

invalidating cache for favoriates

95d66a0

github-actions bot added the platform/frontend AutoGPT Platform - Front end label Sep 29, 2025

github-project-automation bot added this to Frontend Sep 29, 2025

Swiftyos added 2 commits September 29, 2025 10:45

fixed test

b1364b1

fixing tests

b3443e0

Significant-Gravitas deleted a comment from coderabbitai bot Sep 29, 2025

Swiftyos added 2 commits October 1, 2025 10:18

Merge branch 'dev' into swiftyos/caching-pt2

2230c76

fix cache invalidation for agent activity dropdown

ff4b092

Significant-Gravitas deleted a comment from coderabbitai bot Oct 1, 2025

Swiftyos added 2 commits October 1, 2025 11:23

moved cache invalidation earlier in flow to avoid race condition

2f9aba0

add detailed log messages

7afa01a

Significant-Gravitas deleted a comment from coderabbitai bot Oct 1, 2025

invalidate more caches - we have many with very similar names...

c9a7cc6

Swiftyos added 4 commits October 1, 2025 15:07

Merge branch 'dev' into swiftyos/caching-pt2

ced61e2

update caching rules

9f5afff

Merge branch 'swiftyos/caching-pt2' of github.com:Significant-Gravita…

e2c24bd

…s/AutoGPT into swiftyos/caching-pt2

revet over logging in test

e13861a

majdyz previously approved these changes Oct 2, 2025

View reviewed changes

github-project-automation bot moved this from 🆕 Needs initial review to 👍🏼 Mergeable in AutoGPT development kanban Oct 2, 2025

create helper functions for clearing caches

e75cf2b

Swiftyos dismissed majdyz’s stale review via e75cf2b October 2, 2025 09:09

majdyz previously approved these changes Oct 2, 2025

View reviewed changes

Swiftyos added 2 commits October 3, 2025 15:06

Merge branch 'dev' into swiftyos/caching-pt2

7c6a914

fix cache clear helper function

431042a

Swiftyos dismissed majdyz’s stale review via 431042a October 3, 2025 13:10

Swiftyos added 3 commits October 3, 2025 15:21

moved clear cache helper function to cache

eb022e5

added more tests and central place to set page_size to make the cachi…

df1d15f

…ng more robust

fix test

045009a

feat(backend): implement comprehensive caching layer for all GET endpoints (Part 2) #10979

Are you sure you want to change the base?

feat(backend): implement comprehensive caching layer for all GET endpoints (Part 2) #10979

Conversation

Swiftyos commented Sep 24, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Changes

New Cache Modules

Cached Endpoints

Key Features

Test Plan

Performance Impact

Related

Uh oh!

netlify bot commented Sep 24, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

✅ Deploy Preview for auto-gpt-docs-dev canceled.

Uh oh!

netlify bot commented Sep 24, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

✅ Deploy Preview for auto-gpt-docs canceled.

Uh oh!

qodo-merge-pro bot commented Sep 24, 2025

PR Reviewer Guide 🔍

Uh oh!

deepsource-io bot commented Sep 24, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Analysis Summary

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull Request Overview

Reviewed Changes

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Swiftyos commented Sep 26, 2025

Uh oh!

github-actions bot commented Sep 29, 2025

Uh oh!

coderabbitai bot commented Oct 1, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Review skipped

Other AI code review bot(s) detected

Uh oh!

majdyz Oct 2, 2025

Choose a reason for hiding this comment

Uh oh!

majdyz Oct 2, 2025

Choose a reason for hiding this comment

Uh oh!

majdyz Oct 2, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Swiftyos commented Sep 24, 2025 •

edited

Loading

netlify bot commented Sep 24, 2025 •

edited

Loading

netlify bot commented Sep 24, 2025 •

edited

Loading

deepsource-io bot commented Sep 24, 2025 •

edited

Loading

coderabbitai bot commented Oct 1, 2025 •

edited

Loading