Documentation
¶
Index ¶
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
Types ¶
type Executor ¶ added in v0.3.0
Executor runs one specd verb and returns its captured stdout. It is injected by the caller (internal/cmd) so the mcp package never imports the dispatcher — that back-edge would be an import cycle, which is why tool execution lived behind the injection seam rather than inside this package.
type Request ¶ added in v0.3.0
type Request struct {
JSONRPC string `json:"jsonrpc"`
ID any `json:"id,omitempty"`
Method string `json:"method"`
Params json.RawMessage `json:"params,omitempty"`
}
type Response ¶ added in v0.3.0
type Response struct {
JSONRPC string `json:"jsonrpc"`
ID any `json:"id,omitempty"`
Result any `json:"result,omitempty"`
Error *ResponseError `json:"error,omitempty"`
}
type ResponseError ¶ added in v0.3.0
type Tool ¶ added in v0.3.0
type Tool struct {
Name string `json:"name"`
Description string `json:"description"`
InputSchema map[string]any `json:"inputSchema"`
}
func BrainTools ¶ added in v0.3.0
func CoreTools ¶ added in v0.3.0
func CoreTools() []Tool
CoreTools derives the MCP tool palette entirely from core.Command metadata — descriptions and input schemas are generated, never hand-authored, so there is one source of truth and no drift (spec 03 R5, C.8). Flag enums map to JSON Schema `enum` and declared defaults to `default`.
Click to show internal directories.
Click to hide internal directories.