194faac9b7
- 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.
18 lines
621 B
Go
18 lines
621 B
Go
package tools
|
|
|
|
import "embed"
|
|
|
|
// SkillsFS contains the capability-grouped SKILL.md files for injection
|
|
// into sandbox workspaces. Each SKILL teaches the LLM how to use a group
|
|
// of system tools via `tai tool <name>`.
|
|
//
|
|
//go:embed skills
|
|
var SkillsFS embed.FS
|
|
|
|
// SystemPrompt is the shared content appended to both CLAUDE.md and AGENTS.md
|
|
// in sandbox workspaces. It provides environment variable documentation and
|
|
// the `tai tool` calling convention. Stored as a single source file to prevent
|
|
// content drift between the two runner instruction files.
|
|
//
|
|
//go:embed prompts/system-tools.md
|
|
var SystemPrompt []byte
|