Repository navigation
test(e2e): isolate verification instructions by turn - #8
Merged
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Isolate each Forward/Managed live E2E verification turn. Echo turns answer directly, while resource turns read only the requested file, environment variable, skill and memory. Skill fixtures follow the current response format and no longer suppress other requested values.
Why
The previous system prompt broadly instructed tool use on every turn. Global verification spent the echo turn searching for a value already supplied in the message and timed out. The existing internal prompt correction was lost when the mirror reconstructed tests from GitHub main; this PR puts that correction in the source repository.
Verification
make lintmake buildmake testmake docs-checkgo build ./examples/...go test -count=1 -tags live -run '^$' ./forward ./managedcdbdbafcc8403a9f920da85552df21b1cc34e0df), with-count=1and cleanup. The two execution E2Es each exercised echo/SSE, file/environment and skill/memory turns; active cancellation and scoped Run retrieval/listing also passed.Existing marker, tool-use, SSE and cleanup assertions remain in place. Tests are not skipped and timeouts are not increased.
Impact
Only two live test files change. Public SDK behavior, contract fixtures, API reference and package version are unchanged. Changelog/migration notes are not applicable to this test-only correction. Python/TypeScript source changes are not included. No credentials, environment files or generated logs are committed.