FIX: Repair XPIA ai_recruiter workflow (HTTP modalities, PDF stem, Message API)#2227
Open
romanlutz wants to merge 3 commits into
Open
FIX: Repair XPIA ai_recruiter workflow (HTTP modalities, PDF stem, Message API)#2227romanlutz wants to merge 3 commits into
romanlutz wants to merge 3 commits into
Conversation
…iguration HTTPXAPITarget._DEFAULT_CONFIGURATION previously declared only text input, so uploading a file produced by PDFConverter (which emits binary_path) raised "This target supports only the following data types: text". Broaden the default configuration to also declare image_path, audio_path, video_path, and binary_path input modalities, per the framework-sanctioned _DEFAULT_CONFIGURATION mechanism. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> Copilot-Session: 70150e7c-5826-4b3b-95d3-773f247b71b8
…arget-file-modalities
…w to Message API PDFConverter now preserves the source template's filename stem in the output name (behind a numeric uniqueness prefix), so downstream consumers that derive artifact names from the filename retain provenance instead of a purely numeric name. Generated (non-template) PDFs keep the default serializer name. The ai_recruiter XPIA integration test and the 5_workflow doc are migrated to the current Message.from_prompt API, replacing the removed SeedGroup/SeedPrompt entry point for attack_content / processing_prompt. Together with the HTTPXAPITarget input-modality fix, this repairs the end-to-end XPIA ai_recruiter workflow. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> Copilot-Session: 70150e7c-5826-4b3b-95d3-773f247b71b8
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.
Description
This repairs the end-to-end XPIA
ai_recruiterworkflow, which was broken in three independent places. The workflow generates a PDF withPDFConverter, uploads it to an HTTP endpoint viaHTTPXAPITarget, and then triggers candidate search.HTTPXAPITarget input modalities.
_DEFAULT_CONFIGURATIONonly declared text input, so uploading abinary_pathfile emitted byPDFConverterfailed up front withThis target supports only the following data types: text. It now declarestext,image_path,audio_path,video_path, andbinary_path, using the framework-sanctioned_DEFAULT_CONFIGURATIONmechanism (see.github/instructions/targets.instructions.md). No send/upload logic changed; this was purely a missing capability declaration.PDF template stem preservation. When converting an existing template PDF,
PDFConverternow preserves the source file's stem in the output name behind a numeric uniqueness prefix (e.g.<ticks>_Jonathon_Sanchez.pdf), so downstream consumers that derive artifact names from the filename retain provenance instead of a purely numeric name. Generated (non-template) PDFs keep the default numeric serializer name.Message API migration. The
ai_recruiterintegration test and the5_workflowdoc are updated to the currentMessage.from_prompt(...)API, replacing the removedSeedGroup/SeedPromptentry point forattack_content/processing_prompt.Tests and Documentation
test_default_configuration_supports_file_path_input_modalitiesandtest_send_prompt_async_binary_path_upload(HTTPXAPITarget), plustest_existing_pdf_output_preserves_source_stem(PDFConverter). All 28 unit tests across the two affected files pass; ruff / format / ty pre-commit hooks are green.ai_recruiterintegration test to the new Message API (exercised underRUN_ALL_TESTSwith live services).doc/code/executor/5_workflow.pyand its synced.ipynbto the new Message API; jupytext/notebook pre-commit hooks pass. The notebook was not re-executed.