Skip to main content
Glama
anortham

COA Goldfish MCP

by anortham

recall

Retrieve recent or searched working context, session state, and task progress with fuzzy search. Filter by time range, workspace, scope, and memory type for precise results.

Instructions

Enhanced memory recall with fuzzy search support. Can search or just show recent memories. Perfect for "what did I work on yesterday?" questions.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
limitNoMaximum results (default: 10)
queryNoSearch query (optional - if not provided, shows recent memories)
scopeNoSearch scope (default: "current")
sinceNoTime range (default: "7d")
typeNoMemory type filter (optional)
workspaceNoSpecific workspace (optional)

Implementation Reference

  • Core handler function for the 'recall' tool. Performs fuzzy memory search (or recent recall without query), formats results with timestamps, workspaces, content summaries, and tags. Returns structured RecallResponse via buildToolContent.
    async recall(args: {
      query?: string;
      since?: string;
      workspace?: string;
      scope?: 'current' | 'all';
      type?: string;
      tags?: string[];
      limit?: number;
      format?: OutputMode;
    }) {
      const {
        query,
        since = '7d',
        workspace,
        scope = 'current',
        type,
        tags,
        limit = 10,
        format
      } = args;
    
      try {
        let memories;
        
        if (query) {
          // Use fuzzy search
          memories = await this.searchEngine.searchMemories({
            query,
            since,
            workspace,
            scope,
            type,
            tags,
            limit
          });
        } else {
          // Return recent memories
          memories = await this.searchEngine.searchMemories({
            since,
            workspace,
            scope,
            type,
            tags,
            limit
          });
        }
    
        if (memories.length === 0) {
          const searchInfo = query ? ` matching "${query}"` : '';
          const formatted = `🧠 No memories found${searchInfo} in the last ${since}`;
          const data = { query, since, scope, workspace, memoriesFound: 0 } as const;
          return buildToolContent('recall', formatted, data as any, format);
        }
    
        // Build formatted output
        const output = ['🧠 Recent Memories:'];
    
        for (const memory of memories) {
          const age = this.formatAge(memory.timestamp);
          const typeIcon = this.getTypeIcon(memory.type);
          const workspaceInfo = memory.workspace === this.storage.getCurrentWorkspace() 
            ? '' 
            : ` [${memory.workspace}]`;
    
          output.push(`${typeIcon} [${memory.id.slice(-6)}] ${age}${workspaceInfo}`);
          
          if (typeof memory.content === 'object' && memory.content && 'description' in memory.content) {
            const contentObj = memory.content as { description?: string };
            output.push(`   ${contentObj.description}`);
          } else {
            const contentStr = typeof memory.content === 'string' 
              ? memory.content 
              : JSON.stringify(memory.content);
            output.push(`   ${contentStr.slice(0, 200)}${contentStr.length > 200 ? '...' : ''}`);
          }
    
          if (memory.tags && memory.tags.length > 0) {
            output.push(`   Tags: ${memory.tags.join(', ')}`);
          }
    
          output.push('');
        }
    
        const formatted = output.join('\n');
        const data = {
          memoriesFound: memories.length,
          timeRange: since,
          memories: memories.map(m => ({
            id: m.id,
            type: m.type,
            age: this.formatAge(m.timestamp),
            content: typeof m.content === 'string' ? m.content : JSON.stringify(m.content),
            workspace: m.workspace,
            tags: m.tags
          }))
        } as const;
        return buildToolContent('recall', formatted, data as any, format);
      } catch (error) {
        return {
          content: [
            {
              type: 'text',
              text: `❌ Recall failed: ${error instanceof Error ? error.message : String(error)}`
            }
          ]
        };
      }
    }
  • Input schema and metadata for the 'recall' tool, defining optional query, time range, workspace filters, tags, and output format. Provided via SearchTools.getToolSchemas().
    {
      name: 'recall',
      description: 'Restore working context after breaks or /clear. Shows recent activity without query. Use when resuming work sessions.',
      inputSchema: {
        type: 'object',
        properties: {
          query: {
            type: 'string',
            description: 'Search query (optional - if not provided, shows recent memories)'
          },
          since: {
            type: 'string',
            description: 'Time range (default: "7d")',
            default: '7d'
          },
          workspace: {
            type: 'string',
            description: 'Workspace name or path (e.g., "coa-goldfish-mcp" or "C:\\source\\COA Goldfish MCP"). Will be normalized automatically.'
          },
          scope: {
            type: 'string',
            enum: ['current', 'all'],
            description: 'Search scope (default: "current")',
            default: 'current'
          },
          type: {
            type: 'string',
            enum: ['checkpoint'],
            description: 'Content type filter - only checkpoints available (Memory objects deprecated)'
          },
          tags: {
            type: 'array',
            items: { type: 'string' },
            description: 'Filter by exact tags (all tags must match)'
          },
          limit: {
            type: 'number',
            description: 'Maximum results (default: 10)',
            default: 10
          },
          format: {
            type: 'string',
            enum: ['plain', 'emoji', 'json', 'dual'],
            description: 'Output format override (defaults to env GOLDFISH_OUTPUT_MODE or dual)'
          }
        }
      }
    }
  • TypeScript interface defining the structured output response for the 'recall' tool.
    export interface RecallResponse extends FormattedResponse {
      operation: 'recall';
      memoriesFound: number;
      timeRange: string;
      memories?: Array<{
        id: string;
        type: string;
        age: string;
        content: string;
        workspace?: string;
        tags?: string[];
      }>;
    }
  • Tool dispatch registration in the main CallToolRequest handler switch statement, routing 'recall' calls to this.searchTools.recall().
    case 'recall':
      return await this.searchTools.recall(args || {});
  • Tool list registration: includes SearchTools.getToolSchemas() (containing 'recall') in the list of available tools returned by ListToolsRequest.
    this.server.setRequestHandler(ListToolsRequestSchema, async () => {
      return {
        tools: [
          // Unified checkpoint tool (replaces checkpoint, restore_session, and search functionality)
          UnifiedCheckpointTool.getToolSchema(),
          
          // Search and timeline tools
          ...SearchTools.getToolSchemas(),
          
          // Unified TODO tool (replaces create_todo_list, view_todos, update_todo)
          getTodoToolSchema(),
          
          // Plan tool for strategic planning and design decisions
          getPlanToolSchema(),
          
          // Standup tool for intelligent work aggregation and reporting
          getStandupToolSchema(),
          
          getListWorkspacesToolSchema(),
          
          // Intel tool for project knowledge management  
          getIntelToolSchema()
        ],
      };
Behavior2/5

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

No annotations are provided, so the description carries the full burden of behavioral disclosure. It mentions 'enhanced memory recall' and 'fuzzy search support,' which hints at functionality, but lacks critical details: it doesn't specify what 'memories' are (e.g., notes, tasks, sessions), how results are returned (format, ordering), whether there are rate limits, or authentication requirements. For a tool with 6 parameters and no annotations, this is a significant gap.

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 concise and front-loaded: the first sentence states the core functionality, the second explains the two modes, and the third provides a usage example. Each sentence adds value without redundancy. However, it could be slightly more structured by explicitly separating features from use cases.

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

Completeness3/5

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

Given the tool's complexity (6 parameters, no output schema, no annotations), the description is incomplete. It covers the basic purpose and usage but lacks details on behavioral traits (e.g., what 'memories' entail, result format) and doesn't leverage sibling context to clarify differentiation. It's adequate as a minimum viable description but has clear gaps for effective agent use.

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

Parameters3/5

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

Schema description coverage is 100%, so the schema fully documents all 6 parameters. The description adds minimal value beyond the schema: it implies the tool can 'search' (mapping to the 'query' parameter) or 'show recent memories' (hinting at default behavior without 'query'), but doesn't explain parameter interactions or provide additional context. Baseline 3 is appropriate when the schema does the heavy lifting.

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

Purpose4/5

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

The description clearly states the tool's purpose: 'Enhanced memory recall with fuzzy search support. Can search or just show recent memories.' It specifies the verb ('recall') and resource ('memories'), and the example question ('what did I work on yesterday?') provides helpful context. However, it doesn't explicitly differentiate from sibling tools like 'search_history' or 'timeline', which likely have overlapping functionality.

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

Usage Guidelines3/5

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

The description provides implied usage guidance: 'Perfect for "what did I work on yesterday?" questions' suggests it's for retrieving recent personal work memories. It mentions two modes (search vs. show recent) but doesn't explicitly state when to use this tool versus alternatives like 'search_history' or 'timeline', nor does it outline exclusions or prerequisites.

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

Related 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/anortham/coa-goldfish-mcp'

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