registry/tools/validate-examples/main.go
Kalle Bylin 0ffcd36047
Add optional subfolder field to Repository schema (#363)
Adds an optional `subfolder` field to the Repository object in
server.json schema to support MCP servers located within subdirectories
of monorepo structures.

## Motivation and Context

Addresses issue #360 .

The current MCP registry schema doesn't explicitly define how to
reference servers that are located within a subdirectory of a Git
repository (a "monorepo" structure). This is a common pattern (e.g.,
[microsoft/mcp](https://github.com/microsoft/mcp) &
[awslabs/mcp](https://github.com/awslabs/mcp).

## How Has This Been Tested?

- **Validator tests**: Added 6 new test cases covering valid subfolders
and security validation (path traversal, absolute paths, invalid
characters, etc.)
- **Schema validation**: All 11 examples in `examples.md` validate
successfully against both `server.schema.json` and
`registry-schema.json`
- **End-to-end validation**: Publisher tool (`mcp-publisher init`) works
correctly and generates valid templates without subfolder
- **Backward compatibility**: Existing server.json files without
subfolder continue to work unchanged

## Breaking Changes

**None.** This is a fully backward-compatible change:
- The `subfolder` field is optional (uses `omitempty` JSON tag)
- Existing server.json files work unchanged
- Publisher tools generate valid templates without subfolder by default
- All existing validation and schemas remain compatible

## Types of changes
- [ ] Bug fix (non-breaking change which fixes an issue)
- [x] New feature (non-breaking change which adds functionality)
- [ ] Breaking change (fix or feature that would cause existing
functionality to change)
- [x] Documentation update

## Checklist
- [x] I have read the [MCP
Documentation](https://modelcontextprotocol.io)
- [x] My code follows the repository's style guidelines
- [x] New and existing tests pass locally
- [x] I have added appropriate error handling
- [x] I have added or updated documentation as needed


### Implementation Details

**Schema Changes:**
- Added `subfolder` property to Repository definition in
`server.schema.json` and `openapi.yaml`
- Field is optional with description and example (`"src/everything"`)

**Go Model:**
- Added `Subfolder string` field to Repository struct with
`json:"subfolder,omitempty"` tag
- Maintains backward compatibility through omitempty behavior

**Security-Focused Validation:**
- Validates subfolder is a clean relative path (no leading `/`)
- Prevents path traversal attacks (no `..` sequences)
- Blocks malformed paths (no trailing `/`, empty segments, invalid
characters)
- Uses regex validation for allowed characters: `^[a-zA-Z0-9\-_./]+$`

**Documentation:**
- Added monorepo example to `examples.md` demonstrating subfolder usage
- Updated `repository_references.md` with concise subfolder
documentation
- Maintained consistency across all documentation with same example path

**Testing:**
- 6 new integration tests in validator covering security and format
validation
- Updated example count validation tool for new example
- All existing tests continue to pass

### Design Decisions

1. **Optional by design**: Most servers don't need subfolder
(single-repo pattern is common)
2. **Clean separation**: Repository URL stays clean, subfolder is
separate metadata
3. **Security first**: Comprehensive validation prevents common
path-based attacks
4. **Publisher tool unchanged**: Default templates don't include
subfolder to avoid confusion for typical use cases

This enables publishers with monorepo structures to clearly specify
server locations while maintaining simplicity for the common
single-repository case.
2025-09-05 21:29:09 +01:00

211 lines
6.2 KiB
Go

// validate-examples validates JSON examples in docs/server-json/examples.md
// against both schema.json and registry-schema.json.
//
// For more information, see docs/server-json/README.md
package main
import (
"bytes"
"encoding/json"
"fmt"
"log"
"os"
"path/filepath"
"regexp"
"strings"
"github.com/modelcontextprotocol/registry/internal/validators"
apiv0 "github.com/modelcontextprotocol/registry/pkg/api/v0"
jsonschema "github.com/santhosh-tekuri/jsonschema/v5"
)
const (
// expectedExampleCount is the number of JSON examples we expect to find in examples.md
// IMPORTANT: Only change this count if you have intentionally added or removed examples
// from the examples.md file. This check prevents accidental formatting changes from
// causing examples to be skipped during validation.
expectedExampleCount = 11
)
func main() {
log.SetFlags(0) // Remove timestamp from logs
if err := runValidation(); err != nil {
log.Fatalf("Error: %v", err)
}
}
func runValidation() error {
basePath := filepath.Join("docs", "server-json")
examplesPath := filepath.Join(basePath, "examples.md")
schemaPath := filepath.Join(basePath, "server.schema.json")
registrySchemaPath := filepath.Join(basePath, "registry-schema.json")
examples, err := extractExamples(examplesPath)
if err != nil {
return fmt.Errorf("failed to extract examples: %w", err)
}
log.Printf("Found %d examples in examples.md\n", len(examples))
if len(examples) != expectedExampleCount {
return fmt.Errorf("expected %d examples but found %d - if this is intentional, update expectedExampleCount in %s",
expectedExampleCount, len(examples), "tools/validate-examples/main.go")
}
log.Println()
baseSchema, err := compileSchema(schemaPath)
if err != nil {
return fmt.Errorf("failed to compile server.schema.json: %w", err)
}
registrySchema, err := compileSchema(registrySchemaPath)
if err != nil {
return fmt.Errorf("failed to compile registry-schema.json: %w", err)
}
validatedCount := 0
for i, example := range examples {
log.Printf("Example %d:", i+1)
if validateExample(example, baseSchema, registrySchema) {
validatedCount++
}
log.Println()
}
if validatedCount != expectedExampleCount {
return fmt.Errorf("validation failed: expected %d examples to pass all validations but only %d did",
expectedExampleCount, validatedCount)
}
log.Printf("Successfully validated all %d examples!", validatedCount)
return nil
}
func validateExample(ex example, baseSchema, registrySchema *jsonschema.Schema) bool {
var data any
if err := json.Unmarshal([]byte(ex.content), &data); err != nil {
log.Printf(" ❌ Invalid JSON: %v", err)
return false
}
// Extract server portion if this is a PublishRequest format
serverData := data
publishRequestValid := true
if dataMap, ok := data.(map[string]any); ok {
if server, exists := dataMap["server"]; exists {
// This is a PublishRequest format - validate only expected properties exist
for key := range dataMap {
if key != "server" && key != "x-publisher" {
log.Printf(" Invalid PublishRequest property: ❌ %s (only 'server' and optional 'x-publisher' are allowed)", key)
publishRequestValid = false
}
}
serverData = server
}
}
baseValid := validateAgainstSchema(serverData, baseSchema, "server.schema.json")
registryValid := validateAgainstSchema(serverData, registrySchema, "registry-schema.json")
goValidatorValid := validateWithObjectValidator(serverData)
// Only count as validated if all validations passed
return publishRequestValid && baseValid && registryValid && goValidatorValid
}
func validateAgainstSchema(data any, schema *jsonschema.Schema, schemaName string) bool {
if err := schema.Validate(data); err != nil {
log.Printf(" Validating against %s: ❌", schemaName)
log.Printf(" Error: %v", err)
return false
}
log.Printf(" Validating against %s: ✅", schemaName)
return true
}
func validateWithObjectValidator(serverData any) bool {
var serverDetail apiv0.ServerJSON
serverDataBytes, err := json.Marshal(serverData)
if err != nil {
log.Printf(" Validating with Go Validator: ❌")
log.Printf(" Error marshaling server data: %v", err)
return false
}
if err := json.Unmarshal(serverDataBytes, &serverDetail); err != nil {
log.Printf(" Validating with Go Validator: ❌")
log.Printf(" Error unmarshaling to ServerDetail: %v", err)
return false
}
if err := validators.ValidateServerJSON(&serverDetail); err != nil {
log.Printf(" Validating with Go Validator: ❌")
log.Printf(" Error: %v", err)
return false
}
log.Printf(" Validating with Go Validator: ✅")
return true
}
type example struct {
content string
line int
}
func extractExamples(path string) ([]example, error) {
data, err := os.ReadFile(path)
if err != nil {
return nil, err
}
content := string(data)
// Regex to match JSON code blocks in markdown
// Captures everything between ```json and ```
re := regexp.MustCompile("(?s)```json\n(.*?)\n```")
matches := re.FindAllStringSubmatchIndex(content, -1)
var examples []example
for _, match := range matches {
if len(match) < 4 {
// should never happen
return nil, fmt.Errorf("invalid match - expected at least 4 indices but got %d", len(match))
}
start, end := match[2], match[3]
// line numbers start at 1
line := 1 + strings.Count(content[:start], "\n")
examples = append(examples, example{
content: content[start:end],
line: line,
})
}
return examples, nil
}
func compileSchema(path string) (*jsonschema.Schema, error) {
compiler := jsonschema.NewCompiler()
compiler.Draft = jsonschema.Draft7
// For registry-schema.json, we need to register the base schema it references
if strings.Contains(path, "registry-schema.json") {
basePath := filepath.Join(filepath.Dir(path), "server.schema.json")
baseData, err := os.ReadFile(basePath)
if err != nil {
return nil, fmt.Errorf("failed to read base schema: %w", err)
}
// Add the base schema to the compiler with the expected URL
if err := compiler.AddResource("https://static.modelcontextprotocol.io/schemas/2025-07-09/server.schema.json", bytes.NewReader(baseData)); err != nil {
return nil, fmt.Errorf("failed to add base schema resource: %w", err)
}
}
return compiler.Compile(path)
}