Linked API MCP

Instructions

Implementation Reference

• src/tools/fetch-post.ts:7-90 (handler)

  The FetchPostTool class implements the 'fetch_post' tool. It extends OperationTool, sets the tool name to 'fetch_post', maps to the specific LinkedIn API operation, defines the Zod input schema for validation, and provides the MCP Tool definition including JSON inputSchema and description.

  export class FetchPostTool extends OperationTool<TFetchPostParams, unknown> {
    public override readonly name = 'fetch_post';
    public override readonly operationName = OPERATION_NAME.fetchPost;
    protected override readonly schema = z.object({
      postUrl: z.string(),
      retrieveComments: z.boolean().optional(),
      retrieveReactions: z.boolean().optional(),
      commentsRetrievalConfig: z
        .object({
          limit: z.number().min(1).max(500).optional(),
          replies: z.boolean().optional(),
          sort: z.enum(['mostRelevant', 'mostRecent']).optional(),
        })
        .optional(),
      reactionsRetrievalConfig: z
        .object({
          limit: z.number().min(1).max(1000).optional(),
        })
        .optional(),
    });
  
    public override getTool(): Tool {
      return {
        name: this.name,
        description:
          'Open a LinkedIn post and retrieve its data, with optional comments and reactions. (st.openPost action).',
        inputSchema: {
          type: 'object',
          properties: {
            postUrl: {
              type: 'string',
              description:
                "LinkedIn URL of the post. (e.g., 'https://www.linkedin.com/posts/username_activity-id')",
            },
            retrieveComments: {
              type: 'boolean',
              description:
                'Optional. When true, also retrieve comments for the post. Configure via commentsRetrievalConfig.',
            },
            retrieveReactions: {
              type: 'boolean',
              description:
                'Optional. When true, also retrieve reactions for the post. Configure via reactionsRetrievalConfig.',
            },
            commentsRetrievalConfig: {
              type: 'object',
              description:
                'Optional. Applies only when retrieveComments is true. Controls comments retrieval (limit, replies, sort).',
              properties: {
                limit: {
                  type: 'number',
                  description:
                    'Optional. Max number of comments to retrieve. Defaults to 10, with a maximum value of 500.',
                },
                replies: {
                  type: 'boolean',
                  description: 'Optional. When true, include replies to comments (threaded).',
                },
                sort: {
                  type: 'string',
                  enum: ['mostRelevant', 'mostRecent'],
                  description:
                    "Optional. Sort order for comments. One of 'mostRelevant' or 'mostRecent'.",
                },
              },
            },
            reactionsRetrievalConfig: {
              type: 'object',
              description:
                'Optional. Applies only when retrieveReactions is true. Controls reactions retrieval (limit).',
              properties: {
                limit: {
                  type: 'number',
                  description:
                    'Optional. Max number of reactions to retrieve. Defaults to 10, with a maximum value of 1000.',
                },
              },
            },
          },
          required: ['postUrl'],
        },
      };
    }
  }

• src/tools/fetch-post.ts:10-26 (schema)

  Zod schema for validating the input parameters of the fetch_post tool.

  protected override readonly schema = z.object({
    postUrl: z.string(),
    retrieveComments: z.boolean().optional(),
    retrieveReactions: z.boolean().optional(),
    commentsRetrievalConfig: z
      .object({
        limit: z.number().min(1).max(500).optional(),
        replies: z.boolean().optional(),
        sort: z.enum(['mostRelevant', 'mostRecent']).optional(),
      })
      .optional(),
    reactionsRetrievalConfig: z
      .object({
        limit: z.number().min(1).max(1000).optional(),
      })
      .optional(),
  });

• src/linked-api-tools.ts:48-49 (registration)

  FetchPostTool is imported and instantiated as part of the LinkedApiTools collection, which aggregates all LinkedIn API tools.

  new FetchPersonTool(progressCallback),
  new FetchPostTool(progressCallback),

• src/utils/linked-api-tool.ts:36-58 (helper)

  Base OperationTool class that provides the core execution logic for operation-based tools like fetch_post. It locates the corresponding LinkedIn API operation by name and executes it with progress tracking.

  export abstract class OperationTool<TParams, TResult> extends LinkedApiTool<TParams, TResult> {
    public abstract readonly operationName: TOperationName;
  
    public override execute({
      linkedapi,
      args,
      workflowTimeout,
      progressToken,
    }: {
      linkedapi: LinkedApi;
      args: TParams;
      workflowTimeout: number;
      progressToken?: string | number;
    }): Promise<TMappedResponse<TResult>> {
      const operation = linkedapi.operations.find(
        (operation) => operation.operationName === this.operationName,
      )! as Operation<TParams, TResult>;
      return executeWithProgress(this.progressCallback, operation, workflowTimeout, {
        params: args,
        progressToken,
      });
    }
  }

• src/linked-api-server.ts:22-24 (registration)

  LinkedApiMCPServer exposes the tools from LinkedApiTools as MCP Tools via getTools(), effectively registering fetch_post for the MCP server.

  public getTools(): Tool[] {
    return this.tools.tools.map((tool) => tool.getTool());
  }