Cursor
thordata-serp-mcp Skill = Rules for using ThorData SERP with Cursor AI
Cursor can load rule definitions for
thordata-serp-mcpvia its rules mechanism. Once integrated, Cursor AI utilizes this skill to determine the search context, output fields, and processing boundaries when handling tasks related to ThorData SERP, public search results, SEO rankings, news monitoring, shopping prices, or local business data.
This article explains how to convert the existing SKILL.md into a rule file compatible with Cursor. Please refer to the relevant product documentation for details on the ThorData SERP MCP service, API, and authentication configuration.
Rule loading method
Cursor's project rules are typically loaded from the following directories:
.cursor/rules/*.mdc
Meanwhile, thordata-serp-mcp/SKILL.md is the standard file for a Codex skill. Cursor does not automatically load the Codex skill folder as a Rules directory.
Therefore, when using this skill in Cursor, you need to copy the contents of SKILL.md into the Cursor rules file:
thordata-serp-mcp/SKILL.md -> .cursor/rules/thordata-serp-mcp.mdc
The rules are still sourced from the original SKILL.md, which avoids the need to maintain two sets of inconsistent instructions.
Applicable Scenarios
You can integrate this skill when you need to handle the following tasks in Cursor:
• ThorData SERP queries • Public search result analysis • SEO ranking checks • News monitoring • Product price or shopping result analysis • Maps, local business, or local search tasks • Batch SERP query tasks
After configuration, Cursor AI will refer to this skill to make a determination:
• When to use ThorData SERP • How to distinguish between different SERP scenarios • Which fields to retain in the output • How to indicate when the tool is unavailable • Which fields cannot be inferred or filled in
Preparation
Please prepare the following first:
• Cursor installed
• A project requiring integration with these rules
• The thordata-serp-mcp folder
• thordata-serp-mcp/SKILL.md
• Optional: thordata-serp-mcp/references/serp-rules.md
To perform actual searches, you also need to configure the available ThorData SERP MCP tool within Cursor. The skill only provides the rules; it does not handle the configuration of the MCP tool itself.
Project Rules and Global Rules
It is recommended to prioritize using project-specific rules.
Project rules apply only to the current project, making them suitable for testing, team collaboration, and project-based management, as well as easy to delete or adjust later.
Only consider adding rules to the global configuration when multiple Cursor projects require the same set of ThorData SERP rules.
Configuration Steps
Open Project
Open the project that requires ThorData SERP in Cursor.
Create the Rules directory.
Create in the project root directory:
.cursor/rules
If the directory already exists, you can use it directly.
Create a rule file
Create in the .cursor/rules directory:
thordata-serp-mcp.mdc
The complete path should be:
.cursor/rules/thordata-serp-mcp.mdc
Write SKILL.md content
Open:
thordata-serp-mcp/SKILL.md
Copy the contents of SKILL.md to:
.cursor/rules/thordata-serp-mcp.mdc
Once saved, Cursor can use this skill as a rule for the current project.
Additional detailed rules (optional)
references/serp-rules.md is the supplementary rule file for this skill; it contains details on engine/type validation, fallback strategies, batch tasks, error handling, and privacy boundaries.
For more comprehensive behavioral constraints, you can enable:
thordata-serp-mcp/references/serp-rules.md
Append the required content to:
.cursor/rules/thordata-serp-mcp.mdc
For basic usage, simply copying SKILL.md is usually sufficient.
Reload project
After saving the rules file, Cursor typically reads the rules automatically. To ensure the rules take effect, you can close and reopen the current project.
Verify whether the validation rules are in effect
Enter in cursor chat:
Use Thordata SERP to search for the latest news about OpenAI and return the title, source, publication time, and link.
If configured correctly, Cursor AI should recognize this as a ThorData SERP news search scenario and focus on source links, publication times, and query parameters in its responses.
FAQ
Can Cursor read the raw SKILL.md directly?
Relying on Cursor to automatically read the original SKILL.md is not recommended.
Cursor looks for rules in .cursor/rules/*.mdc files. To ensure the rules are applied reliably, you should copy the content of SKILL.md into .cursor/rules/thordata-serp-mcp.mdc.
Why not use simplified rule templates?
SKILL.md serves as the primary source of rules for the skill. Directly reusing SKILL.md minimizes rule divergence and prevents inconsistencies between the simplified rules found in tutorials and the actual skill itself.
Is it necessary to add serp-rules.md?
serp-rules.md is a supplementary rules file and is not mandatory.
It is recommended to integrate SKILL.md first. You should incorporate the relevant content from references/serp-rules.md only when the task requires more granular engine/type validation, fallback rules, batch task rules, or error handling rules.
What should I do if the cursor prompts that it cannot search?
This is typically not a failure to access the rules. The rules simply instruct Cursor AI on how to handle ThorData SERP tasks. Performing actual searches requires the ThorData SERP MCP tool to be configured within Cursor.
How do I restrict this rule to the current project only?
Place the rules in the current project's:
.cursor/rules/thordata-serp-mcp.mdc
This way, the rule will only apply to the current project.
How to uninstall
To remove this Skill rule, simply delete the following file:
.cursor/rules/thordata-serp-mcp.mdc
Deleting the rule file will not delete or modify any MCP tool configurations.
Last updated
Was this helpful?