feat(core): new system prompt configurator for validation feedback. #469
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Outdated suggestions cannot be applied.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.
This pull request introduces a new validation feedback mechanism for AI function calling, along with updates to system prompts and orchestration logic to enhance error handling and correction capabilities. The most important changes include the addition of a specialized validation prompt, integration of the validation prompt into function calling workflows, and documentation updates to clarify the foundational behavior of the AI agent.
Validation Feedback Mechanism:
packages/core/prompts/validate.md
: Added detailed documentation and examples for theIValidation.IFailure
structure, including guidelines for analyzing validation errors, providing correction feedback, and handling common error scenarios such as type mismatches, format violations, and union type failures.System Prompt Updates:
packages/core/src/constants/AgenticaSystemPrompt.ts
: Introduced a newVALIDATE
system prompt to guide AI agents in correcting function calling parameter generation errors. This prompt includes detailed instructions and examples for handling validation failures.Orchestration Logic Enhancements:
packages/core/src/orchestrate/call.ts
: Updated thecorrect
function to use the newVALIDATE
system prompt fromAgenticaSystemPrompt
for error correction during function calls.Documentation Improvements:
packages/core/src/structures/IAgenticaSystemPrompt.ts
: Expanded the description of the common system prompt to emphasize its role in defining the AI agent's core identity and communication style across all interaction phases.