8 posts tagged with "release"

SharpOMatic 10.0.7 is now available.

The headline addition is fallback models for Model Call nodes. A node can now define an ordered list containing its primary model and one or more fallbacks. When a provider is unavailable, rate limited, times out, or encounters another eligible transient failure, SharpOMatic can continue the logical call with the next enabled model. Fallbacks can use a different model, connector, or provider, and the editor highlights capability or configuration differences before the workflow is run.

This release also adds OpenTelemetry tracing for workflow runs, node executions, and model provider calls; repeat counts for evaluation rows; compilation caching for C# scripts and expressions; and support for passing a broader range of file URLs to compatible models. The editor gains clearer navigation, a dedicated string and JSON value viewer, live run duration, and expanded fallback-aware metrics.

Changes​

• Added ordered primary and fallback models to Model Call nodes, with support for multiple fallback entries.
• Added matching model cards in the editor, including capability warnings and controls to add, remove, disable, and re-enable models.
• Added compatibility checks for text, image and file input, tool calling, structured output, and active model parameter values.
• Reused compatible primary Tool Calling and Structured Output settings when executing a fallback while retaining fallback-specific values where settings are incompatible.
• Added conservative default fallback handling for rate limits, provider HTTP 5xx responses, network failures, and timeouts.
• Prevented fallback after visible response output or tool execution has started, avoiding duplicate output and repeated tool side effects.
• Added IEngineNotification.ModelFallbackOverride so hosts can customize whether an eligible failure should use the next model.
• Added provider-specific failure translation for OpenAI, Azure OpenAI, Anthropic, Azure AI Foundry Anthropic, and Google model callers.
• Added automatic in-memory upgrade of version 1 Model Call nodes to the version 2 ordered model representation. Upgraded workflows are persisted on their next normal save.
• Added one model-call metric per provider attempt, with logical call IDs, attempt numbers, normalized failure categories, and provider HTTP status codes.
• Expanded the Metrics page with logical-call totals, fallback usage, recovery rates, final failure rates, fallback-aware charts and breakdowns, and expandable attempt details.
• Added OpenTelemetry spans for workflow runs, node executions, model provider calls, and tool execution using standard activity and GenAI conventions.
• Added workflow, node, model, connector, token, cost, status, conversation, fan-out/fan-in, batch, switch, and Gosub attributes to traces.
• Recorded original exceptions on failed node spans, including exception type, message, and stack trace.
• Added telemetry options for disabling engine tracing or explicitly enabling sensitive prompt, response, and tool content.
• Added a fixed Repeat value to evaluation rows. A row can run from 0 to 10,000 times, with zero skipping that row.
• Added evaluation row repeat support to the editor, execution engine, CSV import/export, transfers, and repository providers.
• Cached compiled C# scripts and expressions for Code nodes, Switch branches, context expressions, and evaluation inputs, avoiding repeated Roslyn compilation across executions.
• Used non-streaming model calls when all incremental AG-UI response output is suppressed, while preserving the configured stored and forwarded events.
• Added URL input support for PDFs, Office documents, spreadsheets, presentations, CSV, JSON, JSONL, HTML, Markdown, plain text, XML, and YAML in addition to images.
• Updated Model Call editor labels from image input to file input where the configured model accepts broader content types.
• Fixed Fan In completion so a retired pre-merge thread cannot supply the workflow's final context.
• Added back buttons to workflow, connector, model, evaluation, and evaluation-run pages.
• Added a full-screen viewer for string context values, including automatic formatted JSON display and clipboard copying.
• Added current run status and duration to the workflow trace panel, with duration updating while the run is active.
• Added Claude Fable 5 metadata for Anthropic and Azure AI Foundry Anthropic.
• Added GPT-5.6 Luna, Sol, and Terra metadata for OpenAI and Azure OpenAI.

SharpOMatic 10.0.6 is now available.

This release expands provider coverage with Anthropic and Azure AI Foundry Anthropic support. It also improves day-to-day editor workflows with workflow folders, evaluation row import/export, stronger transfer behavior, auth-gate hooks, and more control over what model-call tool activity is streamed to AG-UI clients.

Under the hood, this release tightens model-call behavior across streaming, tool calls, image inputs, metadata lookups, and database migrations, with matching documentation and test updates.

Changes​

• Added Anthropic and Azure AI Foundry Anthropic connector configs, model configs, and model caller services.
• Added Claude Haiku, Sonnet, and Opus metadata for Anthropic and Azure AI Foundry Anthropic models.
• Added workflow folders across the repository, editor APIs, Angular workflows UI, AG-UI workflow selection, transfers, migrations, and docs.
• Added CSV import and export for evaluation rows.
• Added a delete-all action for evaluation rows.
• Added editor auth-gate support with front-end auth provider, token, and interceptor plumbing.
• Added generic sign-in messaging for authenticated editor hosts.
• Added per-tool AG-UI output settings for Model Call nodes so tool activity can be streamed, hidden, or controlled individually.
• Added Model Call tool-exit support with context-aware exit results and cleaner graceful termination behavior.
• Preserved tool-call result text in chat history replay.
• Improved Model Call streaming replay for partial function calls and tool-call results.
• Added URL image input support for model calls.
• Allowed null entries in model-call image input paths.
• Added StreamEventHelper access from registered tool methods.
• Added repository metadata lookup optimizations and focused metadata tests.
• Added SQL Server migration-table and transaction compatibility fixes.
• Cleaned up the SQL Server 10.0.6 migration path.
• Increased parallel node execution capacity to 50 nodes.
• Updated workflow run and model-call metrics to retain readable snapshot names instead of GUID-only references.
• Fixed designer delete-key handling.
• Fixed editor notification settings.
• Improved embedded asset relative path handling.
• Fixed missing Azure OpenAI deployment-name metadata.
• Clarified the additional license terms.

SharpOMatic 10.0.5 is now available.

The headline addition is a Metrics page covering both model call usage and workflow run health. Each model call node execution writes a metric row capturing the workflow, connector, model, token usage, cost, duration, and any failure details. Workflow run metrics are written when a run reaches a terminal state and include an aggregated snapshot of the model usage for that run. Both sets of metrics are retained independently of run history, so the data remains available after detailed run records are pruned.

The AG-UI stream viewer has also been updated: the tabs previously labelled "Stream" are now labelled "AG-UI", and each event kind is displayed using its canonical AG-UI protocol name (for example TEXT_MESSAGE_START, TOOL_CALL_ARGS, STEP_FINISHED) rather than a friendly paraphrase.

Changes​

• Added a Metrics page with model call and workflow run dashboards.
• Model call metrics record workflow, node, connector, model, duration, token usage, cost, and failure details per execution.
• Workflow run metrics record run status, duration, conversation turn details, failure details, and aggregated model call usage per terminal run.
• Both metric types use snapshot names so data remains available after workflows, connectors, or models are renamed or deleted.
• Renamed the AG-UI stream viewer tabs from "Stream" to "AG-UI".
• Updated stream event labels to use canonical AG-UI protocol event names.
• Removed the conversation lease mechanism, simplifying the conversation model.
• Added GPT-5.5 model configs for OpenAI and Azure OpenAI.
• Added Google Gemini 3.1 Flash Lite and Gemini 3.5 Flash model configs.

SharpOMatic 10.0.3 is now available.

Big addition is support for conversations, allowing multi-turn workflows. This enables a chatbot and agent style interactions and with the additionof AG-UI it becomes a via backend provider to front end chat clients. A 'react' based sample called 'sharpy' is provided so you can experiment with this.

Changes​

• Added conversations with suspend and resume support.
• Added AG-UI support using POST with server-sent events.
• Sharpy sample using CopilotKit to talk to workflows.
• Fixed a Google tool calling bug.
• Allow failed evaluations to be deleted.
• Allow grader output to use any numeric type.
• Added options for the overall evaluation result.
• Propagate serialization errors at the end of node execution.

SharpOMatic 10.0.4 is now available.

This release builds on the conversation and AG-UI work from 10.0.3. AG-UI clients can now reload conversation history after a browser refresh or application restart, workflow authors can emit template-based stream messages without writing code, and model-call streaming has more control over what appears in the live stream.

The release also refreshes the Sharpy sample, updates the built-in sample workflows, adds model-caller extension hooks, and brings the major .NET, OpenAI, Azure OpenAI, Google GenAI, Roslyn, and EF Core dependencies forward.

Changes​

• Added an AG-UI conversation history POST endpoint at /history.
• Added optional maxMessages support for history reloads.
• Updated the Sharpy client sample.
• Added the Event Template node for emitting text or reasoning stream messages from templates.
• Added recursive template expansion for context paths and text assets.
• Added model override hooks for OpenAI, Azure OpenAI, and Google GenAI clients.
• Updated the OpenAI and Azure OpenAI callers to use the newer Responses client path.
• Updated Google GenAI integration for the newer SDK API.

SharpOMatic 10.0.2 is now available.

Changes​

• Added asset folders for easier grouping of assets
• Added latest Google Gemini and OpenAI models
• Improved evaluation run output
• Show thinking and tool call traces
• Fixed multiple bug fixes

SharpOMatic 10.0.1 is now available.

Changes​

• Added a Batch node for processing lists in chunks, including parallel batch execution.
• Added a Gosub node so a workflow can call another workflow as a child.
• Added Evaluations with config-driven rows, columns, and graders.
• Added Google AI connector support, including Gemini model usage.
• Added both SQL Server and SQLite Entity Framework providers.
• Improved trace visibility, including thread and parent/child trace context.

I am excited to announce SharpOMatic 10.0.0, our first release. The version number matches the .NET framework version we target: .NET 10.0.

What is included​

• A workflow engine built for AI-focused orchestration.
• An embedded editor UI for designing, running, and debugging workflows.
• An ASP.NET Core host for local development and integration.
• Documentation covering setup and core concepts.

What next​

The first release has minimal functionality, like all first releases. I will be interating quickly to flesh out out the functionality. If there are any requests then just raises an issue on GitHub.