- Introduced `image_generate` tool for generating images from text prompts, with options for specifying output file paths and image dimensions.
- Updated `image_read` functionality to allow optional provider specification for enhanced image analysis.
- Implemented new `GenerateImage` method in the LLM API for seamless integration of image generation capabilities.
- Enhanced documentation to include detailed usage examples for both image reading and generation tools.
- Updated tests to validate new image generation features and ensure robust functionality across image tools.
- Implemented dynamic SMTP provider resolution based on user/team context, improving email sending capabilities.
- Updated the Send and SendT methods to utilize identity information from the context for dynamic provider selection.
- Refactored cloud encryption and decryption methods to delegate to the setting package, streamlining cryptographic operations.
- Enhanced team invitation email sending to include identity context, ensuring accurate user/team information is used.
- Removed obsolete crypto helper functions, simplifying the codebase and improving maintainability.
- Added support for processing image content in the agent, providing placeholder text when image parsing fails.
- Implemented a fallback mechanism to utilize vision tools for image analysis, improving the handling of image content.
- Updated tests to validate the new image handling behavior, ensuring correct responses when vision support is unavailable.
- Enhanced system prompts to guide users on using the new `image_read` tool for image analysis, improving user experience.
- Refactored the initSandboxV2 function to return a structured result, consolidating the runner, computer, configuration, cleanup function, loading message ID, and roles into a single return type.
- Updated the Stream method to utilize the new sandboxV2InitResult structure, improving clarity and reducing complexity in handling sandbox initialization.
- Introduced role management enhancements, allowing for pre-resolved role connectors to be passed through the request, streamlining connector resolution during execution.
- Adjusted various components to support the new roles structure, ensuring consistent handling across the agent's sandbox operations.
- Added logging for connector resolution and role management, improving diagnostics and traceability during sandbox execution.
- Updated .gitignore to include tools/TOOL-REGISTRATION.md for better project organization.
- Skipped tests related to KB/DB search functionality due to temporary unavailability.
- Updated test cases in chat_test.go, search_auth_integration_test.go, search_auto_full_test.go, and others to reflect this change.
- Adjusted search handling in search.go to limit search types to "web" only until KB/DB search is re-enabled.
- Added support for tool loop processing when tool call responses are present and sandbox mode is disabled, improving the assistant's ability to handle complex tool interactions.
- Implemented fallback delegation to a loop fallback mechanism in case of tool loop execution failure, enhancing robustness in error scenarios.
- Updated the assistant message structure to include reasoning content, providing better context for generated responses.
- Enhanced error logging in tool call execution to include detailed content, improving diagnostics for tool call failures.
- Updated system configuration to include a new loop fallback agent, expanding the assistant's capabilities.
- Updated the builtinSearch function to delegate search operations to the websearch tool, improving modularity and maintainability.
- Enhanced context handling by passing the agent context to builtinSearch, allowing for user and team identification during searches.
- Added a new DecryptValue function in cloud.go to streamline value decryption, delegating to the existing config.DecryptValue method.
- Updated .gitignore to include tools/README.md for better project organization.
- Added support for embedding and image generation capabilities in the capabilitiesFromMap function.
- Updated the ToMap method to delegate to the canonical Capabilities.ToMap() method, simplifying the conversion process.
- Enhanced the capabilities handling in various components, including filters for non-chat models in OpenAPI settings.
- Introduced utility functions for filter checking and non-chat model identification, improving overall capabilities management.
- Introduced a new "Heavy" role in the system configuration, allowing for complex reasoning tasks.
- Updated the initialization and environment resolution functions to handle the new Heavy role appropriately.
- Enhanced tests to validate the integration of the Heavy role, ensuring proper connector resolution and model handling.
- Adjusted OpenAPI settings to reflect the addition of the Heavy role, improving overall system capabilities.
- Integrated logging for connector resolution failures, providing clearer diagnostics during fallback scenarios.
- Simplified connector ID extraction in various components, ensuring accurate handling of model identifiers.
- Updated model configuration to support new parameters and capabilities, improving overall provider management.
- Enhanced OpenAPI settings to reflect changes in model options and connector behavior, ensuring better alignment with upstream API requirements.
- Updated the timeout parameter in the TestProcessCall_Timeout_Short test to 1 second to ensure it accurately reflects expected behavior during LLM round-trip execution.
- This change aims to improve the reliability of the test by aligning the timeout with realistic processing times.
- Updated the system configuration to include new role-level defaults for Light, Vision, and Audio connectors.
- Refactored the resolveSystemConnector function to prioritize per-agent overrides, improving connector resolution logic.
- Enhanced LLMConnector integration across various components to streamline settings retrieval and capabilities management.
- Improved error handling and logging for connector-related operations, ensuring better diagnostics and user feedback.
- Replaced 'Voice' with 'Audio' in the system configuration and related tests to better reflect functionality.
- Introduced new methods for role management in the llmprovider, allowing for dynamic retrieval of roles based on user and team context.
- Updated the OpenAPI settings to support new role management endpoints and capabilities.
- Enhanced the handling of API keys in provider management, allowing for optional plain-text retrieval.
- Updated logger methods to respect the new `config.Silent` flag, allowing suppression of console output in development mode.
- Refactored `buildTurnResponse` to utilize a shared `buildTrace` function for improved tool call association.
- Introduced JSON output capabilities in `OutputWriter` for detailed trace and duration reporting.
- Enhanced reporting mechanisms to include trace details in JSON and Markdown formats for better diagnostics.
- Added support for a new `Trace` structure to capture detailed execution information, including tool calls and completion data.
- Introduced `shellQuoteForPlatform` to build shell-safe command strings tailored for Windows and POSIX environments.
- Updated `buildSandboxEnvPrompt` to display the correct environment variable syntax based on the operating system.
- Refactored the command copying logic in the runner to use platform-aware commands for copying custom OpenCode tools.
- Removed outdated test scenarios related to the vision connector, streamlining the test suite.
- Implemented two new test scenarios: one to verify that read.ts is not copied when no vision connector is present, and another to confirm that it is copied when a vision connector is configured.
- Updated the runner logic to conditionally copy custom tools into the OpenCode global config directory based on the presence of a vision connector in the configuration.
- Registered new "opencode" and "opencode/cli" runners in the sandbox initialization.
- Updated the `runExecStep` function to use the computer's working directory and set the HOME environment variable.
- Refactored attachment processing in the Claude package to delegate to shared functionality, improving code maintainability.
- Updated testConnectorID comment and model version in TestAnthropicStreamRetry to reflect the latest model (Claude Haiku 4.5).
- Added Test command to the SUI command set with associated flags for improved testing capabilities.
- Updated the spdystream dependency version from v0.5.0 to v0.5.1 in go.mod and go.sum files.
- Modified testConnectorID comment to reflect the updated model version (Claude Haiku 3.5).
- Updated model version in TestAnthropicStreamRetry to use "claude-3-5-haiku-20241022".
Implement yao doc process list/inspect/validate and yao doc runtime
list/inspect/validate commands. Validate uses engine addressing logic
(process.Of) and checks dynamic-ID group registries (model, store, fs,
task, schedule) to verify resources actually exist.
- cmd/doc/: CLI command tree with process and runtime subcommands
- cmd/root.go: wire docCmd into rootCmd
- 27 process doc.yml + doc.go pairs across yao packages
- cmd/doc/doc_test.go: integration tests
Made-with: Cursor
- Added Size field to represent the file size in bytes.
- Introduced ContentType field to specify the MIME type of the attachment.
- Updated API endpoint to support bundling multiple files into a ZIP archive.
- Upgraded several dependencies in go.mod and go.sum, including golang.org/x/crypto, golang.org/x/net, golang.org/x/sys, and google.golang.org/grpc to their latest versions for improved security and performance.
- Updated opentelemetry packages to v1.43.0 and added new indirect dependencies for better observability support.
- Adjusted the handling of execute messages in stream.go to ensure proper persistence based on message status.
- Enhanced command handling in command.go to maintain system prompt constraints during continuation turns.
- Replaced chunkRecord with recordedEvent to simplify event recording.
- Introduced activeToolID to manage the currently streaming tool, allowing for better state handling during parsing.
- Enhanced closeStreamingTool and suspendStreamingTool methods for improved tool message management.
- Updated parsing logic to handle multiple concurrent tool calls more effectively.
- Added utility functions for extracting message groups and properties from recorded events.
- Implemented GetWorkspaceID method in the context package to retrieve the workspace ID.
- Updated buildContextVariables to include WORKSPACE_ID in the context variables if available.
- Enhanced buildEnv function to set CTX_WORKSPACE_ID in the environment configuration based on the workspace ID.
- Refactored role connector handling to support multiple connectors in the environment setup.
- Added tests for new role connector functionality and workspace ID integration.
- Introduced Vision and Voice fields in the SystemConfig and System types to support new capabilities.
- Updated resolveEnvStrings function to handle environment variables for Vision and Voice.
- Enhanced unit tests to validate the new Vision and Voice configurations, ensuring correct environment variable resolution.
- Added GetVisionConnector and GetVoiceConnector functions to retrieve connectors for vision and voice capabilities.
- Upgraded AWS SDK for Go v2 and its related packages in go.mod and go.sum to improve compatibility and access to new features.
- Updated session management logic in the Runner to ensure proper handling of chat sessions.
- Added new unit tests for keyword search functionality in the Assistant model, enhancing search capabilities across locales.
- Refactored JSON handling in the Assistant package to streamline options configuration.
- Introduced NanoToTime and TimeToNano functions for converting between UnixNano and time.Time, improving clarity and consistency in timestamp management.
- Updated Assistant model to utilize the new utility functions for created_at and updated_at fields.
- Refactored Xun store methods to ensure UTC compatibility for timestamp serialization across different database drivers.
- Added unit tests for the new conversion functions to validate their correctness and behavior.
- Updated GetAssistants and DeleteAssistants methods to utilize jsonContainsValue for filtering tags and locales, improving query compatibility across database drivers.
- Renamed related test functions to reflect the new JSON handling approach.
- Removed deprecated jsonLikeExpr and whereJsonLike methods to streamline codebase.
- Introduced utility functions for applying LIKE conditions on JSON columns, enhancing compatibility with PostgreSQL.
- Simplified the filtering logic for tags and keywords in the GetAssistants and DeleteAssistants methods.
- Added unit tests for the new JSON handling functions to ensure correctness across different database drivers.
- Updated database configuration to include PostgreSQL 14.0 in CI workflows.
- Enhanced JSON null detection in the Xun store to support PostgreSQL dialect.
- Refactored time conversion utilities to handle multiple database formats, including PostgreSQL.
- Improved cleanup logic in the DataCleaner to ensure proper order of operations for referential integrity.
- Added utility functions for converting between Unix timestamps and database timestamps.
Removed the resolveEnvRef function and replaced its usage with str.EnvVar for processing environment variables in BuildCreateOptions and buildEnv functions. This change simplifies the code and enhances consistency in how environment variables are managed across the sandbox.
Updated the resolveEnvRef function to handle $ENV_B64.XXX references, allowing for base64-decoding of environment variables. This enhancement is particularly useful for managing multi-line values such as SSH private keys, improving the flexibility of environment variable handling.
Modified the initSandboxV2 function to return a per-request copy of the SandboxConfig, ensuring that concurrent requests to the same assistant have their own mutable configuration. Updated related logging and cleanup processes to utilize the new configuration structure. Additionally, introduced a DefaultOneShotMaxAge constant to manage the lifecycle of OneShot containers more effectively.
cfg (ast.SandboxV2) is a pointer shared across all concurrent calls to
the same Assistant. GetComputer writes cfg.ID = <random hex> for each
oneshot request, so concurrent tasks overwrite each other's ID.
When cleanup runs in defer, cfg.ID holds whichever task last wrote it —
not the ID of the box this request created — so manager.Remove/Heartbeat
silently fails with "sandbox: not found" and the container leaks.
Fix: derive the box ID from computer.ComputerInfo().BoxID, which is the
immutable b.id set at Create time and never mutated by concurrent callers.
Made-with: Cursor
Three related bugs caused oneshot containers to run indefinitely:
1. watcher.go: switch b.policy had no case OneShot, so even when the
idle timeout fired, no remove action was emitted.
2. watcher.go + manager.go/recoverBoxes: idleTimeoutD was only set for
Session and LongRunning on recovery; OneShot defaulted to 0, which
caused the watcher to hit the `timeout <= 0` early-return and skip
all checks entirely.
3. agent/sandbox/v2/options.go: same gap — opts.IdleTimeout == 0 guard
only filled defaults for Session and LongRunning.
Fix: add DefaultOneShotIdleTimeout (30 min), wire it in recoverBoxes
and options.go, and add case OneShot → Remove in watcher.go.
Made-with: Cursor
- Introduced a new `workspace` field across various robot-related structures, including `CreateRobotRequest`, `UpdateRobotRequest`, and `RobotResponse`, allowing for better organization and management of robots within specific workspaces.
- Updated database queries and response mappings to accommodate the new workspace field, ensuring seamless integration with existing functionalities.
- Enhanced agent execution context to include workspace information, improving the contextual awareness of agents during operations.
- Added tests to validate the creation and updating of robots with workspace data, ensuring robust functionality and backward compatibility.
- Modified the buildEnv function to set environment variables for ANTHROPIC_MODEL and related keys dynamically based on the provided model value, enhancing flexibility and configurability.
- Added detailed logging of environment variables passed to the session in the Stream method, improving visibility into the execution context.
- Implemented conditional logging to display sensitive variables in a secure manner, ensuring that only relevant information is shown.
- Renamed ClaudeRunner to Runner for consistency across the codebase, aligning with the sandbox Runner interface.
- Updated buildCommand and Stream methods to utilize AssistantID directly from StreamRequest, improving clarity and reducing dependency on Config.
- Modified tests to reflect the changes in runner instantiation and argument handling, ensuring compatibility with the new Runner structure.
- Enhanced environment variable setup in buildEnv to include WORKDIR, streamlining the execution context for the runner.
- Renamed `buildInput` to `buildLastUserMessageJSONL` to better reflect its purpose of constructing JSONL from the last user message.
- Updated test cases to use the new function, ensuring they accurately verify the behavior of skipping system messages and only including the last user message.
- Removed the now redundant `buildFirstRequestJSONL` function, streamlining the input building process in the Claude command tests.
- Updated logging in various methods to include detailed traces for better monitoring of execution flow, including assistantID and chatID.
- Improved context management by ensuring proper release of resources and tracking of execution durations in agent calls.
- Added heartbeat logging in the Claude parser to monitor stream processing and prevent potential issues with long-running tasks.
- Refined the .gitignore file to exclude additional markdown files in the sandbox directory.
- Modified the buildAgentContext function to accept assistantID as a parameter, enhancing the context building process for agent calls.
- Updated multiple Call methods to pass the assistantID, ensuring proper context management during agent interactions.
- Introduced session ID and name handling in the buildArgs function, allowing for better tracking of chat sessions.
- Added chatIDToSessionUUID and sanitizeSessionName functions to generate and format session identifiers.
- Updated the Stream method to store session information and manage session lifecycle more effectively.
- Implemented KillSessionCmd for precise process termination based on session names in both Windows and POSIX platforms.
- Enhanced tests to cover new session management features and ensure correct behavior in various scenarios.
- Updated agent context to ensure locale is propagated correctly, defaulting to "en" when no locale is provided. This change enhances internationalization support and ensures consistent behavior across different contexts.
- Adjusted test assertions to reflect the updated execution trigger logic for times mode robots, allowing for a maximum of 2 triggers instead of 1.
- Modified integration tests to utilize the new dry run executor in the manager configuration, ensuring consistent execution behavior during testing.
- Updated multiple test cases to replace direct manager instantiation with the new configuration method, enhancing test reliability and maintainability.
- Increased the timeout duration from 60 seconds to 120 seconds for waiting on execution statuses in multiple E2E test cases, ensuring more reliable test outcomes during longer execution scenarios.
- Implemented pre-acquisition of execution slots in the Tick method to prevent race conditions, ensuring that robots do not submit duplicate executions.
- Updated TryAcquireSlot method to support idempotent behavior, allowing for early slot reservation without consuming additional resources.
- Modified worker execution logic to skip pre-checks for robots that have already acquired a slot, streamlining the execution process.
- Improved error handling during execution submission to ensure proper tracking and removal of failed executions.
- Enhanced the extractSandboxVersion function to support multiple input types, including *sandboxTypes.SandboxConfig and map[string]any, for better flexibility in version retrieval.
- Simplified the version extraction process, ensuring consistent handling of sandbox configurations.
- Added support for sandbox configuration in the LoadPath function, allowing for better management of sandbox settings.
- Updated tests to validate the retrieval of assistant tags with type filters, ensuring accurate responses for both assistant and robot types.
- Enhanced the assistant model to differentiate between sandbox versions, improving the handling of sandbox data in the database.
- Improved API responses to include computer filter details for V2 sandboxes, enhancing the information returned to clients.
