Skip to main content
Glama
DollhouseMCP

DollhouseMCP

Official

portfolio_element_manager

Manage individual portfolio elements between local storage and GitHub repository. Download or upload personas, skills, and other elements using fuzzy matching for easy identification.

Instructions

Manage individual elements between your local portfolio and GitHub repository. USE THIS TO DOWNLOAD/UPLOAD INDIVIDUAL PERSONAS, SKILLS, OR OTHER ELEMENTS! When a user asks to 'download X persona from my GitHub' or 'upload Y skill to GitHub', use this tool. Operations: 'download' (GitHub→local), 'upload' (local→GitHub), 'list-remote' (see what's on GitHub), 'compare' (diff local vs GitHub). FUZZY MATCHING enabled - 'verbose victorian scholar' will find 'Verbose-Victorian-Scholar'. After downloading, use reload_elements then activate_element.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
operationYesThe operation to perform. 'download' = get from GitHub to local, 'upload' = send from local to GitHub, 'list-remote' = see what's on GitHub, 'compare' = see differences
element_nameNoName of the element (required for download, upload, compare). FUZZY MATCHING ENABLED: Just type the name naturally - 'verbose victorian scholar', 'Victorian Scholar', 'verbose-victorian', etc. will all work
element_typeNoType of element (required for download, upload, compare). For personas use 'personas'
filterNoFilters for bulk operations
optionsNoAdditional options. For downloads, use options:{force:true} to skip confirmations

Implementation Reference

  • Tool registration and schema definition for 'portfolio_element_manager'. Delegates execution to server.handleSyncOperation.
      tool: {
        name: "portfolio_element_manager",
        description: "Manage individual elements between your local portfolio and GitHub repository. USE THIS TO DOWNLOAD/UPLOAD INDIVIDUAL PERSONAS, SKILLS, OR OTHER ELEMENTS! When a user asks to 'download X persona from my GitHub' or 'upload Y skill to GitHub', use this tool. Operations: 'download' (GitHub→local), 'upload' (local→GitHub), 'list-remote' (see what's on GitHub), 'compare' (diff local vs GitHub). FUZZY MATCHING enabled - 'verbose victorian scholar' will find 'Verbose-Victorian-Scholar'. After downloading, use reload_elements then activate_element.",
        inputSchema: {
          type: "object",
          properties: {
            operation: {
              type: "string",
              enum: ["list-remote", "download", "upload", "compare"],
              description: "The operation to perform. 'download' = get from GitHub to local, 'upload' = send from local to GitHub, 'list-remote' = see what's on GitHub, 'compare' = see differences"
            },
            element_name: {
              type: "string",
              description: "Name of the element (required for download, upload, compare). FUZZY MATCHING ENABLED: Just type the name naturally - 'verbose victorian scholar', 'Victorian Scholar', 'verbose-victorian', etc. will all work"
            },
            element_type: {
              type: "string",
              enum: ["personas", "skills", "templates", "agents", "memories", "ensembles"],
              description: "Type of element (required for download, upload, compare). For personas use 'personas'"
            },
            filter: {
              type: "object",
              properties: {
                type: {
                  type: "string",
                  enum: ["personas", "skills", "templates", "agents", "memories", "ensembles"],
                  description: "Filter by element type"
                },
                author: {
                  type: "string",
                  description: "Filter by author username"
                },
                updated_after: {
                  type: "string",
                  description: "Filter by update date (ISO 8601 format)"
                }
              },
              description: "Filters for bulk operations"
            },
            options: {
              type: "object",
              properties: {
                force: {
                  type: "boolean",
                  description: "Force overwrite existing elements. Use force:true when downloading to skip confirmation prompts"
                },
                dry_run: {
                  type: "boolean",
                  description: "Preview changes without applying them"
                },
                include_private: {
                  type: "boolean",
                  description: "Include elements marked as private/local-only"
                }
              },
              description: "Additional options. For downloads, use options:{force:true} to skip confirmations"
            }
          },
          required: ["operation"]
        }
      },
      handler: (args: any) => server.handleSyncOperation(args)
    }
  • Direct handler implementation (server.handleSyncOperation). Processes args, calls PortfolioSyncManager, formats results.
    async handleSyncOperation(options: SyncOperationOptions, indicator: string = '') {
      try {
        await this.configManager.initialize();
        
        // Check if sync is enabled (allow list-remote and compare even when disabled)
        const syncEnabled = this.configManager.getSetting('sync.enabled');
        const readOnlyOperations = ['list-remote', 'compare'];
        if (!syncEnabled && !readOnlyOperations.includes(options.operation)) {
          return {
            content: [{
              type: "text",
              text: `${indicator}⚠️ **Sync is Disabled**\n\n` +
                    `Portfolio sync is currently disabled for privacy.\n\n` +
                    `To enable sync:\n` +
                    `\`dollhouse_config action: "set", setting: "sync.enabled", value: true\`\n\n` +
                    `You can still use \`list-remote\` and \`compare\` to view differences.`
            }]
          };
        }
        
        // Map our operation to PortfolioSyncManager's SyncOperation format
        const syncOp: SyncOperation = {
          operation: this.mapOperation(options.operation),
          element_name: options.element_name,
          element_type: options.element_type || options.filter?.type, // Use filter.type if element_type not provided
          bulk: options.operation.includes('bulk'),
          show_diff: options.operation === 'compare',
          force: options.options?.force,
          confirm: options.options?.force || options.options?.dry_run === false // force implies confirm, dry_run=false means confirm
        };
        
        // Call the unified handleSyncOperation method
        const result = await this.syncManager.handleSyncOperation(syncOp);
        
        // Format the result based on the operation type
        return this.formatResult(result, options, indicator);
        
      } catch (error) {
        const sanitizedError = SecureErrorHandler.sanitizeError(error);
        return {
          content: [{
            type: "text",
            text: `${indicator}❌ Sync operation failed: ${sanitizedError.message}`
          }]
        };
      }
    }
  • Core handler dispatching to specific operations (download, upload, list-remote, compare) with full business logic.
    public async handleSyncOperation(params: SyncOperation): Promise<SyncResult> {
      try {
        // Check if sync is enabled in config
        const config = this.configManager.getConfig();
        if (!config.sync.enabled && params.operation !== 'list-remote') {
          return {
            success: false,
            message: 'Sync is disabled. Enable it with: dollhouse_config --action update --setting sync.enabled --value true'
          };
        }
        
        // Check bulk permissions
        if (params.bulk) {
          const bulkAllowed = this.isBulkOperationAllowed(params.operation, config);
          if (!bulkAllowed.allowed) {
            return {
              success: false,
              message: bulkAllowed.message
            };
          }
        }
        
        // Handle operations
        switch (params.operation) {
          case 'list-remote':
            return await this.listRemoteElements(params.element_type);
            
          case 'download':
            if (params.bulk) {
              return await this.bulkDownload(params.element_type, params.confirm);
            } else if (params.element_name) {
              return await this.downloadElement(
                params.element_name,
                params.element_type!,
                params.version,
                params.force
              );
            } else {
              return {
                success: false,
                message: 'Element name required for individual download'
              };
            }
            
          case 'upload':
            if (params.bulk) {
              return await this.bulkUpload(params.element_type, params.confirm);
            } else if (params.element_name) {
              return await this.uploadElement(
                params.element_name,
                params.element_type!,
                params.confirm
              );
            } else {
              return {
                success: false,
                message: 'Element name required for individual upload'
              };
            }
            
          case 'compare':
            if (params.element_name && params.element_type) {
              return await this.compareVersions(
                params.element_name,
                params.element_type,
                params.show_diff
              );
            } else {
              return {
                success: false,
                message: 'Element name and type required for comparison'
              };
            }
            
          default:
            return {
              success: false,
              message: `Unknown operation: ${params.operation}`
            };
        }
      } catch (error) {
        logger.error('Sync operation failed', {
          operation: params.operation,
          error: error instanceof Error ? error.message : String(error)
        });
        
        return {
          success: false,
          message: `Sync operation failed: ${error instanceof Error ? error.message : String(error)}`
        };
      }
    }
  • Key helper method for downloading individual elements from GitHub with fuzzy matching and conflict resolution.
    private async downloadElement(
      elementName: string,
      elementType: ElementType,
      version?: string,
      force?: boolean
    ): Promise<SyncResult> {
      try {
        const config = this.configManager.getConfig();
        
        // Validate element name
        const validation = UnicodeValidator.normalize(elementName);
        if (!validation.isValid) {
          return {
            success: false,
            message: `Invalid element name: ${validation.detectedIssues?.[0] || 'unknown error'}`
          };
        }
        
        // Get token and set it
        const token = await TokenManager.getGitHubTokenAsync();
        if (!token) {
          return {
            success: false,
            message: 'GitHub authentication required'
          };
        }
        
        this.repoManager.setToken(token);
        
        // Get GitHub index
        const index = await this.indexer.getIndex();
        
        // Find the element - first try exact match, then fuzzy match
        const entries = index.elements.get(elementType) || [];
        let entry = entries.find(e => e.name === elementName);
        
        // If exact match not found, try fuzzy matching
        if (!entry) {
          // Try case-insensitive exact match first
          entry = entries.find(e => e.name.toLowerCase() === elementName.toLowerCase());
          
          // If still not found, try fuzzy matching
          if (!entry) {
            const fuzzyMatch = this.findFuzzyMatch(elementName, entries);
            if (fuzzyMatch) {
              logger.info(`Fuzzy match found: '${elementName}' matched to '${fuzzyMatch.name}'`);
              entry = fuzzyMatch;
            }
          }
        }
        
        if (!entry) {
          // Generate helpful suggestions
          const suggestions = this.getSuggestions(elementName, entries);
          const suggestionText = suggestions.length > 0 
            ? `\n\nDid you mean one of these?\n${suggestions.map(s => `  • ${s.name}`).join('\n')}`
            : '';
          
          return {
            success: false,
            message: `Element '${elementName}' (${elementType}) not found in GitHub portfolio${suggestionText}`
          };
        }
        
        // Check for local conflicts
        const localPath = this.portfolioManager.getElementPath(elementType, `${elementName}.md`);
        let hasLocalVersion = false;
        let localContent: string | null = null;
        
        try {
          localContent = await fs.readFile(localPath, 'utf-8');
          hasLocalVersion = true;
        } catch {
          // No local version exists
        }
        
        // Download the element
        const response = await fetch(entry.downloadUrl, {
          headers: {
            'Authorization': `Bearer ${token}`,
            'Accept': 'application/vnd.github.v3.raw'
          }
        });
        
        if (!response.ok) {
          throw new Error(`Failed to download: ${response.statusText}`);
        }
        
        const remoteContent = await response.text();
        
        // Validate content security
        const validationResult = ContentValidator.validateAndSanitize(remoteContent);
        if (!validationResult.isValid && validationResult.severity === 'critical') {
          return {
            success: false,
            message: `Security issue detected in remote content: ${validationResult.detectedPatterns?.join(', ')}`
          };
        }
        
        // Check if content is different
        if (hasLocalVersion && localContent) {
          const localHash = createHash('sha256').update(localContent).digest('hex');
          const remoteHash = createHash('sha256').update(remoteContent).digest('hex');
          
          if (localHash === remoteHash) {
            return {
              success: true,
              message: `Element '${elementName}' is already up to date`
            };
          }
          
          // Show confirmation for overwrite unless force flag is set
          if (config.sync.individual.require_confirmation && !force) {
            const diff = await this.generateDiff(localContent, remoteContent);
            
            return {
              success: false,
              message: `Local version exists. Please confirm download will overwrite:\n\n${diff}\n\nTo proceed, use --force flag`,
              data: { requiresConfirmation: true }
            };
          }
        }
        
        // Save the element
        await fs.mkdir(path.dirname(localPath), { recursive: true });
        await fs.writeFile(localPath, remoteContent, 'utf-8');
        
        logger.info('Element downloaded from GitHub', {
          element: elementName,
          type: elementType,
          version: entry.version
        });
        
        return {
          success: true,
          message: `Successfully downloaded '${elementName}' (${elementType}) from GitHub portfolio`
        };
        
      } catch (error) {
        return {
          success: false,
          message: `Failed to download element: ${error instanceof Error ? error.message : String(error)}`
        };
      }
    }
  • Key helper method for uploading individual elements to GitHub with security checks.
    private async uploadElement(
      elementName: string,
      elementType: ElementType,
      confirm?: boolean
    ): Promise<SyncResult> {
      try {
        const config = this.configManager.getConfig();
        
        // Check for local element
        const localPath = this.portfolioManager.getElementPath(elementType, `${elementName}.md`);
        
        let content: string;
        try {
          content = await fs.readFile(localPath, 'utf-8');
        } catch {
          return {
            success: false,
            message: `Element '${elementName}' (${elementType}) not found locally`
          };
        }
        
        // Check privacy metadata
        const parsed = SecureYamlParser.parse(content, {
          maxYamlSize: 64 * 1024,
          validateContent: false,
          validateFields: false
        });
        
        if (parsed.data?.privacy?.local_only === true) {
          return {
            success: false,
            message: `Element '${elementName}' is marked as local-only and cannot be uploaded`
          };
        }
        
        // Validate content security
        const validationResult = ContentValidator.validateAndSanitize(content);
        if (!validationResult.isValid && validationResult.severity === 'critical') {
          return {
            success: false,
            message: `Security issue detected: ${validationResult.detectedPatterns?.join(', ')}`
          };
        }
        
        // Scan for sensitive content if configured
        if (config.sync.privacy.scan_for_secrets) {
          logger.debug('Scanning for secrets before upload');
          // Implement actual secret scanning
          const secretPatterns = [
            /api[_-]?key\s*[:=]\s*['"][^'"]+['"]/gi,
            /secret\s*[:=]\s*['"][^'"]+['"]/gi,
            /password\s*[:=]\s*['"][^'"]+['"]/gi,
            /token\s*[:=]\s*['"][^'"]+['"]/gi,
            /private[_-]?key\s*[:=]\s*['"][^'"]+['"]/gi
          ];
          
          for (const pattern of secretPatterns) {
            if (pattern.test(content)) {
              return {
                success: false,
                message: `Potential secret detected in content. Please review and remove sensitive information before uploading.`
              };
            }
          }
        }
        
        // Get confirmation if required (unless already confirmed)
        if (config.sync.individual.require_confirmation && !confirm) {
          return {
            success: false,
            message: `Please confirm upload of '${elementName}' (${elementType}) to GitHub.\n\nContent preview:\n${content.substring(0, 500)}...\n\nTo proceed, use --confirm flag`,
            data: { requiresConfirmation: true }
          };
        }
        
        // Get token and validate
        const token = await TokenManager.getGitHubTokenAsync();
        if (!token) {
          return {
            success: false,
            message: 'GitHub authentication required'
          };
        }
        
        // Create a PortfolioElement for the adapter (fixes Issue #913)
        // Using PortfolioElementAdapter instead of incomplete IElement implementation
        const portfolioElement = {
          type: elementType,
          metadata: {
            name: elementName,
            description: parsed.data?.description || '',
            author: parsed.data?.author || 'unknown',
            created: parsed.data?.created || new Date().toISOString(),
            updated: new Date().toISOString(),
            version: parsed.data?.version || '1.0.0',
            tags: parsed.data?.tags || []
          },
          content: content
        };
        
        // Use PortfolioElementAdapter to properly implement IElement interface
        const adapter = new PortfolioElementAdapter(portfolioElement);
        
        // Use PortfolioRepoManager to upload
        this.repoManager.setToken(token);
        
        // DEBUG: Log upload attempt
        logger.debug('[BULK_SYNC_DEBUG] Upload element attempt', {
          elementName,
          elementType,
          hasToken: !!token,
          tokenPrefix: token ? token.substring(0, 10) + '...' : 'none',
          adapterHasMetadata: !!(adapter && adapter.metadata),
          timestamp: new Date().toISOString()
        });
        
        try {
          const url = await this.repoManager.saveElement(adapter, true); // consent is true since we've already checked
          
          logger.info('Element uploaded to GitHub', {
            element: elementName,
            type: elementType,
            url
          });
          
          return {
            success: true,
            message: `Successfully uploaded '${elementName}' (${elementType}) to GitHub portfolio`,
            data: { url }
          };
        } catch (uploadError) {
          // Handle specific errors
          if (uploadError instanceof Error && uploadError.message.includes('repository does not exist')) {
            return {
              success: false,
              message: `GitHub portfolio repository not found. Please initialize it first using init_portfolio tool.`
            };
          }
          throw uploadError;
        }
        
      } catch (error) {
        return {
          success: false,
          message: `Failed to upload element: ${error instanceof Error ? error.message : String(error)}`
        };
      }
    }
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries full burden and does well by disclosing key behavioral traits: it explains the four operations, mentions fuzzy matching functionality, describes what happens after download (need to reload and activate), and implies this is for GitHub synchronization. However, it doesn't mention authentication requirements, rate limits, or error handling.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is appropriately sized and front-loaded with the core purpose in the first sentence. Each subsequent sentence adds value: usage examples, operation explanations, fuzzy matching details, and post-operation steps. Some redundancy exists between the description and schema (operation explanations), but overall it's efficient.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a complex tool with 5 parameters, nested objects, no annotations, and no output schema, the description does well by covering operations, fuzzy matching, and workflow context. It explains what the tool does and when to use it, though it could better address the filter parameter and provide more detail on return values given the lack of output schema.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the baseline is 3. The description adds meaningful context beyond the schema by explaining the purpose of each operation ('download' = GitHub→local, 'upload' = local→GitHub), emphasizing fuzzy matching with examples, and providing usage examples that clarify parameter combinations. However, it doesn't fully explain the filter parameter's purpose.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's purpose with specific verbs ('manage', 'download', 'upload', 'list-remote', 'compare') and resources ('individual elements', 'local portfolio', 'GitHub repository'). It explicitly distinguishes this tool from siblings by specifying it's for individual element operations rather than bulk operations, and provides concrete examples of when to use it.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides explicit guidance on when to use this tool ('download X persona from my GitHub', 'upload Y skill to GitHub') and distinguishes it from alternatives by specifying it's for individual elements. It also provides post-operation guidance ('After downloading, use reload_elements then activate_element'), creating clear workflow context.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/DollhouseMCP/mcp-server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server