- SCHEMA
- TESTS
- CODE
- HUMAN
Short answer
How do I keep my AI's API knowledge up to date?
Your agent uses the Assertible MCP to fetch the latest OpenAPI or Swagger files from your service. This ensures its understanding of your endpoints and schemas matches your actual code. You'll see how this works below.
The results.
Where the data lands.
What happens after your agent syncs your specifications.
SCHEMA
Fresh endpoint maps
The agent pulls the latest definitions. It now knows exactly which paths and methods are active in your service.
TESTS
Accurate test logic
Your agent updates its internal model of your API. This prevents it from suggesting tests for endpoints that no longer exist.
CODE
Valid client generation
When you ask for code, the agent uses the current schema. You get snippets that actually work with your live API.
HUMAN
Manual oversight needed
The AI updates its knowledge, but it won't touch your source code. You still handle the actual implementation and deployment.
The workflow
What your AI does when a spec changes.
The AI handles the heavy lifting of reading and interpreting your documentation.
Fetch files
The agent reaches out to Assertible to grab the current OpenAPI or Swagger files.
sync_specificationParse definitions
The AI reads the raw files to find new endpoints or modified parameters.
sync_specificationUpdate context
The agent adds these new details to its active memory for your current session.
sync_specificationVerify schema
The AI confirms the new structure matches your requirements before answering your next prompt.
sync_specification
Try it
Copy these to start.
Use these prompts to trigger a sync or verify your API knowledge.
Starting points
These are opening lines. Swap the bracketed text for your specific service or endpoint names.
Assertible Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_j2b1fGA9QZp77KBXLw34HQRPgprjB0X3YhgDhJPG/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Assertible capabilities are ready to use.
{
"mcpServers": {
"assertible-mcp": {
"url": "https://edge.vinkius.com/vk_preview_j2b1fGA9QZp77KBXLw34HQRPgprjB0X3YhgDhJPG/mcp"
}
}
}Sync the latest OpenAPI spec from [Service Name] so you have the current endpoint list.
Check the Swagger file for [Service Name] and tell me if any required parameters have changed.
Pull the latest specification from Assertible and write a Python client for the [Endpoint Name] endpoint.
Compare my current API knowledge with the latest spec from [Service Name] to find breaking changes.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Assertible once, then ask.
Click the link to connect your account. Your credentials stay encrypted on our side, and you can immediately start asking your agent to sync your specs.
Connect Assertible to your AIFAQ
How the sync behaves
- 01
Will the AI change my API code automatically?
No. The AI only reads the specification files to update its own understanding. It cannot modify your actual service or codebase.
- 02
Does the AI delete old API definitions?
No. The sync_specification tool updates the AI's context with the new data. It doesn't delete anything in your Assertible account.
- 03
Can the AI fix broken endpoints?
No. The AI can identify that an endpoint has changed based on the spec, but it cannot fix the underlying code.
- 04
How often does the AI check for updates?
The AI only checks for updates when you specifically ask it to sync or when you prompt it to use the latest specification.
- 05
What file formats does it support?
The tool works with OpenAPI and Swagger files provided by your Assertible service.
More questions about Assertible? See everything the Assertible Connector can do
Connect Assertible to Claude, Cursor, ChatGPT & more
