Skip to main content
Glama
0xjcf
by 0xjcf

update-codebase-insight

Modify existing codebase insights by updating content, type, related files, or tags to maintain accurate development documentation and analysis.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
insightIdYes
insightContentNo
insightTypeNo
relatedFilesNo
tagsNo

Implementation Reference

  • MCP tool handler for 'update-codebase-insight'. Calls the updateMemory helper to update the insight in the database and returns a success or error response.
      async ({ insightId, insightContent, insightType, relatedFiles, tags }) => {
        try {
          await updateMemory({
            id: insightId,
            insightContent,
            insightType,
            relatedFiles,
            tags
          });
          
          return {
            content: [{
              type: "text",
              text: `Successfully updated insight with ID: ${insightId}`
            }]
          };
        } catch (error) {
          return {
            content: [{
              type: "text",
              text: `Error updating insight: ${(error as Error).message}`
            }],
            isError: true
          };
        }
      }
    );
  • Input schema for the 'update-codebase-insight' tool using Zod validation.
    {
      insightId: z.number(),
      insightContent: z.string().optional(),
      insightType: z.string().optional(),
      relatedFiles: z.array(z.string()).optional(),
      tags: z.array(z.string()).optional()
    },
  • Registration of the 'update-codebase-insight' tool on the MCP server.
    server.tool(
      "update-codebase-insight",
  • Helper function that performs the actual database update for an insight, including optional updates to content, type, related files, and tags using SQLite transactions.
    export async function updateMemory(updates: {
      id: number;
      insightContent?: string;
      insightType?: string;
      relatedFiles?: string[];
      tags?: string[];
    }): Promise<void> {
      if (!db) {
        db = await createDatabase("memory");
      }
      
      if (!db) {
        throw new Error("Database not initialized");
      }
      
      const { id, insightContent, insightType, relatedFiles, tags } = updates;
      
      // Begin transaction
      await db.exec('BEGIN TRANSACTION');
      
      try {
        // Update basic insight data if provided
        if (insightContent || insightType) {
          let sql = `UPDATE insights SET`;
          const params: any[] = [];
          
          if (insightContent) {
            sql += ` insightContent = ?,`;
            params.push(insightContent);
          }
          
          if (insightType) {
            sql += ` insightType = ?,`;
            params.push(insightType);
          }
          
          // Remove trailing comma
          sql = sql.slice(0, -1);
          
          sql += ` WHERE id = ?`;
          params.push(id);
          
          // Execute update
          await db.run(sql, params);
        }
        
        // Update related files if provided
        if (relatedFiles) {
          // Delete existing relationships
          await db.run(`DELETE FROM relatedFiles WHERE insightId = ?`, [id]);
          
          // Insert new relationships
          for (const filePath of relatedFiles) {
            await db.run(
              `INSERT INTO relatedFiles (insightId, filePath) VALUES (?, ?)`,
              [id, filePath]
            );
          }
        }
        
        // Update tags if provided
        if (tags) {
          // Delete existing tags
          await db.run(`DELETE FROM tags WHERE insightId = ?`, [id]);
          
          // Insert new tags
          for (const tag of tags) {
            await db.run(
              `INSERT INTO tags (insightId, tag) VALUES (?, ?)`,
              [id, tag]
            );
          }
        }
        
        await db.exec('COMMIT');
      } catch (error) {
        await db.exec('ROLLBACK');
        throw error;
      }
    }

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/0xjcf/MCP_CodeAnalysis'

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