[dotnet-port-api] Port skill detailed script errors option - #395
Conversation
Add ContextProviderOptions.IncludeDetailedErrors for skills so script execution failures can opt in to model-visible error details while propagating runner errors by default. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
There was a problem hiding this comment.
Pull request overview
This PR ports the .NET “detailed skill script errors” behavior into the Go skills context provider by introducing an opt-in flag that controls whether script execution failures are exposed to the model vs. propagated as tool-call errors to the caller/middleware.
Changes:
- Added
skills.ContextProviderOptions.IncludeDetailedErrorsto control model-visible script execution error details. - Updated
run_skill_scripttool behavior to propagate script runner errors by default (unless detailed errors are enabled). - Added provider tests covering default error propagation and opt-in detailed error strings; updated the .NET/Go feature comparison doc.
Reviewed changes
Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.
| File | Description |
|---|---|
agent/skills/provider.go |
Adds IncludeDetailedErrors and changes run_skill_script to return (any, error) so runner errors can propagate by default. |
agent/skills/provider_test.go |
Adds tests for default propagation of script-runner errors and opt-in detailed error output. |
docs/dotnet-go-sdk-feature-comparison.md |
Documents the new Go option and its parity with the .NET API. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
This comment has been minimized.
This comment has been minimized.
|
@copilot address the PR feedback |
Co-authored-by: michelle-clayton-work <262183035+michelle-clayton-work@users.noreply.github.com>
Cross-repo consistency reviewResult: ✅ Aligned with .NET — no parity issues found. .NET parity check
Python parity checkPython's
|
Add ContextProviderOptions.IncludeDetailedErrors for skills so script execution failures can opt in to model-visible error details while propagating runner errors by default.