fix(advanced): tell the model only one eval may be in flight - #1081
Merged
Merged
Conversation
There was a problem hiding this comment.
🟢 Approval recommended
The scoped change includes regression coverage and no blocking issues were identified.
Pull request overview
Updates the advanced code interpreter guidance to prevent overlapping eval calls.
Changes:
- Documents the single in-flight
evallimitation. - Recommends
Promise.allfor in-call parallelism. - Adds regression coverage and bumps the version to 0.18.1.
File summaries
| File | Summary |
|---|---|
uv.lock |
Synchronizes the package version. |
tests/agent/advanced/test_code_interpreter.py |
Verifies the rendered guidance. |
src/uipath_langchain/agent/advanced/code_interpreter.py |
Adds the concurrency note to the tool description. |
pyproject.toml |
Updates the package version. |
Review details
- Files reviewed: 3/4 changed files
- Comments generated: 0
- Review effort level: Lite
💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
18 tasks
radu-mocanu
force-pushed
the
fix/eval-single-in-flight-note
branch
from
September 15, 2026 10:02
a5ca178 to
4ce889e
Compare
robert-ursu
approved these changes
Sep 15, 2026
radu-mocanu
force-pushed
the
fix/eval-single-in-flight-note
branch
from
September 15, 2026 11:33
4ce889e to
b98e088
Compare
|
radu-mocanu
added a commit
that referenced
this pull request
Sep 16, 2026
(cherry picked from commit 25acdcc)
This was referenced Sep 16, 2026
radu-mocanu
added a commit
that referenced
this pull request
Sep 17, 2026
(cherry picked from commit 25acdcc)
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
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
evaldescription now says only one call may run at a time, and points atPromise.allfor parallelism inside a single callWhy
the REPL is one interpreter with shared state, so overlapping calls are rejected rather than queued. nothing told the model that: the description is rendered upstream and does not mention it, and per-tool parallelism is not expressible in a tool schema since
parallel_tool_callsis request-level. a model that batches twoevalcalls in one turn loses one toConcurrentEvalErrorand burns a turn recovering.this is a local workaround. the note belongs in the upstream description and can be deleted once it lands there.