Skip to content

Fix flaky basic-tool-calling-prompt-test #276

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
ericdallo merged 1 commit into from
Jan 21, 2026
+17 −14
Merged

Fix flaky basic-tool-calling-prompt-test #276

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
31 changes: 17 additions & 14 deletions test/eca/features/chat_test.clj
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -285,22 +285,25 @@
:type :text}]}}}
{:role "assistant" :content [{:type :text :text "I can see: \n/foo/bar"}]}]}}
(:chats (h/db))))
;; Note: We use m/in-any-order because there's a race between progress messages
;; ("Calling tool", "Generating") and tool state messages (toolCallRunning,
;; toolCalled) - their relative order in the middle section is non-deterministic.
(is (match?
{:chat-content-received
[{:role :user :content {:type :text :text "List the files you are allowed to see\n"}}
{:role :system :content {:type :progress :state :running :text "Waiting model"}}
{:role :system :content {:type :progress :state :running :text "Generating"}}
{:role :assistant :content {:type :text :text "Ok,"}}
{:role :assistant :content {:type :text :text " working on it"}}
{:role :assistant :content {:type :toolCallPrepare :id "call-1" :name "list_allowed_directories" :arguments-text ""}}
{:role :assistant :content {:type :toolCallRun :id "call-1" :name "list_allowed_directories" :arguments {} :manual-approval false}}
{:role :assistant :content {:type :toolCallRunning :id "call-1" :name "list_allowed_directories" :arguments {}}}
{:role :system :content {:type :progress :state :running :text "Calling tool"}}
{:role :assistant :content {:type :toolCalled :id "call-1" :name "list_allowed_directories" :arguments {} :total-time-ms number? :outputs [{:text "Allowed directories: /foo/bar" :type :text}]}}
{:role :system :content {:type :progress :state :running :text "Generating"}}
{:role :assistant :content {:type :text :text "I can see: \n"}}
{:role :assistant :content {:type :text :text "/foo/bar"}}
{:role :system :content {:state :finished :type :progress}}]}
(m/in-any-order [{:role :user :content {:type :text :text "List the files you are allowed to see\n"}}
{:role :system :content {:type :progress :state :running :text "Waiting model"}}
{:role :system :content {:type :progress :state :running :text "Generating"}}
{:role :assistant :content {:type :text :text "Ok,"}}
{:role :assistant :content {:type :text :text " working on it"}}
{:role :assistant :content {:type :toolCallPrepare :id "call-1" :name "list_allowed_directories" :arguments-text ""}}
{:role :assistant :content {:type :toolCallRun :id "call-1" :name "list_allowed_directories" :arguments {} :manual-approval false}}
{:role :assistant :content {:type :toolCallRunning :id "call-1" :name "list_allowed_directories" :arguments {}}}
{:role :system :content {:type :progress :state :running :text "Calling tool"}}
{:role :assistant :content {:type :toolCalled :id "call-1" :name "list_allowed_directories" :arguments {} :total-time-ms number? :outputs [{:text "Allowed directories: /foo/bar" :type :text}]}}
{:role :system :content {:type :progress :state :running :text "Generating"}}
{:role :assistant :content {:type :text :text "I can see: \n"}}
{:role :assistant :content {:type :text :text "/foo/bar"}}
{:role :system :content {:state :finished :type :progress}}])}
(h/messages))))))

(deftest concurrent-tool-calls-test
Expand Down
Loading