[Agent Builder] Initial API tests with mocked LLM #234985
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
sorenlouv
force-pushed
the
one-chat-api-test
branch
2 times, most recently
from
bd2855f to
5de7dc4
Compare
sorenlouv
force-pushed
the
one-chat-api-test
branch
from
6ad3862 to
214bfa5
Compare
…urce-definitions/scripts/fix-location-collection.ts'
sorenlouv
force-pushed
the
one-chat-api-test
branch
from
ef74183 to
a874030
Compare
sorenlouv
added
release_note:skip
sorenlouv
changed the title
pgayvallet
approved these changes
Sep 15, 2025
x-pack/platform/packages/shared/onechat/onechat-common/tools/tool_result.ts
Outdated
Show resolved
Hide resolved
x-pack/solutions/chat/plugins/workchat-app/server/services/chat/generate_conversation_title.ts
Outdated
Show resolved
Hide resolved
x-pack/platform/test/onechat_api_integration/utils/llm_proxy/scenarios.ts
Outdated
Show resolved
Hide resolved
jbudz
approved these changes
Sep 15, 2025
Member
jbudz
left a comment
jbudz
left a comment
There was a problem hiding this comment.
.buildkite/ftr_platform_stateful_configs.yml
csr
approved these changes
Sep 16, 2025
Contributor
csr
left a comment
csr
left a comment
There was a problem hiding this comment.
x-pack/platform/test/tsconfig.json changes LGTM 👍
joemcelroy
approved these changes
Sep 16, 2025
Contributor
💛 Build succeeded, but was flaky
Failed CI StepsTest Failures
Metrics [docs]
History
|
sorenlouv
added a commit
that referenced
this pull request
Sep 19, 2025
A spec was left out in #234985.
sorenlouv
added a commit
that referenced
this pull request
Sep 22, 2025
Follow up to #234985 The chart type was hardcoded to `Line`. This is removed and instead Lens will auto-detect the most likely chart type
CAWilson94
pushed a commit
to CAWilson94/kibana
that referenced
this pull request
Sep 24, 2025
Closes elastic/search-team#10970 This PR adds two API tests for the Agent Builder converse endpoint: - Simple conversation test: Verifies basic conversation functionality, including title generation and response handling - ESQL query test: Tests complex tool-calling flows including esql generation, searching ES and returning a structured tool result to the LLM Instead of calling a real LLM these tests call a simulated LLM (called "LLM Proxy"). It's a lightweight HTTP server that simulates LLM responses. The proxy: - Acts as a drop-in replacement for an actual LLM - Intercepts requests to the LLM and returns predefined responses based on pattern matching - Supports complex conversation flows including tool calls - Enables testing the full chain of interactions between Agent Builder and the LLM This approach allows us to test the integration between Agent Builder and LLMs in a deterministic, cheap and fast way. --------- Co-authored-by: kibanamachine <[email protected]>
CAWilson94
pushed a commit
to CAWilson94/kibana
that referenced
this pull request
Sep 24, 2025
…c#235757) A spec was left out in elastic#234985.
CAWilson94
pushed a commit
to CAWilson94/kibana
that referenced
this pull request
Sep 24, 2025
Follow up to elastic#234985 The chart type was hardcoded to `Line`. This is removed and instead Lens will auto-detect the most likely chart type
niros1
pushed a commit
that referenced
this pull request
Sep 30, 2025
Closes elastic/search-team#10970 This PR adds two API tests for the Agent Builder converse endpoint: - Simple conversation test: Verifies basic conversation functionality, including title generation and response handling - ESQL query test: Tests complex tool-calling flows including esql generation, searching ES and returning a structured tool result to the LLM Instead of calling a real LLM these tests call a simulated LLM (called "LLM Proxy"). It's a lightweight HTTP server that simulates LLM responses. The proxy: - Acts as a drop-in replacement for an actual LLM - Intercepts requests to the LLM and returns predefined responses based on pattern matching - Supports complex conversation flows including tool calls - Enables testing the full chain of interactions between Agent Builder and the LLM This approach allows us to test the integration between Agent Builder and LLMs in a deterministic, cheap and fast way. --------- Co-authored-by: kibanamachine <[email protected]>
niros1
pushed a commit
that referenced
this pull request
Sep 30, 2025
A spec was left out in #234985.
niros1
pushed a commit
that referenced
this pull request
Sep 30, 2025
Follow up to #234985 The chart type was hardcoded to `Line`. This is removed and instead Lens will auto-detect the most likely chart type
rylnd
pushed a commit
to rylnd/kibana
that referenced
this pull request
Oct 17, 2025
Closes elastic/search-team#10970 This PR adds two API tests for the Agent Builder converse endpoint: - Simple conversation test: Verifies basic conversation functionality, including title generation and response handling - ESQL query test: Tests complex tool-calling flows including esql generation, searching ES and returning a structured tool result to the LLM Instead of calling a real LLM these tests call a simulated LLM (called "LLM Proxy"). It's a lightweight HTTP server that simulates LLM responses. The proxy: - Acts as a drop-in replacement for an actual LLM - Intercepts requests to the LLM and returns predefined responses based on pattern matching - Supports complex conversation flows including tool calls - Enables testing the full chain of interactions between Agent Builder and the LLM This approach allows us to test the integration between Agent Builder and LLMs in a deterministic, cheap and fast way. --------- Co-authored-by: kibanamachine <[email protected]>
rylnd
pushed a commit
to rylnd/kibana
that referenced
this pull request
Oct 17, 2025
…c#235757) A spec was left out in elastic#234985.
rylnd
pushed a commit
to rylnd/kibana
that referenced
this pull request
Oct 17, 2025
Follow up to elastic#234985 The chart type was hardcoded to `Line`. This is removed and instead Lens will auto-detect the most likely chart type
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
7 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Closes https://github.com/elastic/search-team/issues/10970
This PR adds two API tests for the Agent Builder converse endpoint:
Instead of calling a real LLM these tests call a simulated LLM (called "LLM Proxy"). It's a lightweight HTTP server that simulates LLM responses. The proxy:
This approach allows us to test the integration between Agent Builder and LLMs in a deterministic, cheap and fast way.