OIP MCP explanation
What the concept is
OIP stands for Object Invocation Protocol. It is a system for interacting with digital objects using simple web requests. In this build, directory rows are the objects. You can invoke these objects by sending data to a specific web address.
MCP stands for Model Context Protocol. It is an open standard. MCP allows an AI model to connect to an MCP server over a session. This server then exposes tools, resources, and prompts that the model can call. MCP is NOT a content-management system. It focuses on providing a rich, stateful environment for AI interactions.
The main difference lies in how they handle interaction. OIP uses plain Uniform Resource Locators (URLs) and receipts. A URL is a web address that identifies a resource on the internet. OIP has no persistent session. Any model that can open a URL can act. MCP, however, relies on a persistent session to maintain context and state for the AI model.
Why this build cares about it
This build, miscsubjects.com, implements OIP. It provides a simple, stateless way for any program or AI model to interact with its objects. This means interactions are independent. Each request is complete on its own. This makes OIP highly scalable and easy to integrate with existing web tools. It aligns with the "tap-and-go" interaction model described in /a/oip-tap-go.
We care about MCP because it is a related protocol for AI interaction. However, OIP offers a simpler, stateless alternative for many use cases. OIP is ideal for one-off actions or integrations where a continuous session state is not needed. It allows for direct, auditable interactions without the overhead of managing a session. This build uses OIP to ensure transparency and simplicity in object invocation.
How to see or use it live with curl against https://miscsubjects.com
You can invoke an OIP object using curl. curl is a command-line tool for transferring data with URLs. An API (Application Programming Interface) is a set of rules that lets programs talk to each other. An endpoint is a specific URL where an API can be accessed. A server is a computer program that provides services to other computer programs.
To invoke an object, you send a request to the /api/dispatch endpoint. You can use either a POST or GET request. The key identifies the object you want to invoke. The body contains the data you send to that object. JSON (JavaScript Object Notation) is a lightweight data-interchange format.
Here is an example using POST:
curl -X POST https://miscsubjects.com/api/dispatch \
-H "Content-Type: application/json" \
-d '{"key": "example-key", "body": {"message": "hello from OIP"}}'Here is an example using GET:
curl "https://miscsubjects.com/api/dispatch?invoke=example-key&body=%7B%22message%22%3A%22hello%20from%20OIP%22%7D"These commands send data to an object identified by example-key. This triggers the object's defined action. You can learn more about OIP API interactions at /a/oip-api and curl usage at /a/oip-curl.
How it relates to MCP when relevant
Both OIP and MCP aim to enable AI models to interact with external systems. However, their approaches differ significantly. OIP is stateless. It uses simple HTTP requests and responses. Each OIP invocation is a standalone event. It does not rely on a continuous connection.
MCP, in contrast, is session-based. An AI model connects to an MCP server and maintains a session. This session provides a persistent context. It allows the model to access tools, resources, and prompts over time. This makes MCP suitable for complex, multi-step AI workflows where state needs to be maintained across interactions.
OIP is simpler for direct, one-off actions. It is easy to integrate with any system that can make an HTTP request. MCP offers a richer, more controlled environment for AI models. It provides a structured way for models to interact with a defined set of capabilities. For a more detailed comparison, see /a/oip-mcp-comparison.
Where the proof lives
Every OIP invocation creates a record. This record is stored in an append-only ledger. An append-only ledger is a data store where new entries are added to the end. Existing entries cannot be changed. This ensures a complete and immutable history of all actions.
After each invocation, a receipt is generated. This receipt confirms that the invocation happened. It also provides unique details about the transaction. You can retrieve a receipt using its unique invocation ID. For example, if an invocation has an ID like inv_12345:
curl "https://miscsubjects.com/api/dispatch?receipt=inv_12345"This ledger and the associated receipts provide an auditable trail. They prove that an object was invoked. They also show what data was sent. This transparency is a core feature of OIP. You can find more details about the ledger and receipts at /a/oip-ledger-receipts.