Metadata-Version: 2.4
Name: osmp-mcp
Version: 1.0.17
Summary: OSMP MCP Server — Agentic AI instruction encoding. 86.8% byte reduction vs JSON. Inference-free decode. Any channel.
Project-URL: Homepage, https://octid.io
Project-URL: Repository, https://github.com/octid-io/cloudless-sky
Project-URL: Documentation, https://github.com/octid-io/cloudless-sky/blob/main/docs/SAL-efficiency-analysis.md
Author-email: Clay Holberg <clay@octid.io>
License: Apache License 2.0 — see https://github.com/octid-io/cloudless-sky/blob/main/LICENSE
License-File: LICENSE
Keywords: agentic,ai,encoding,lora,mcp,mesh,osmp,protocol,semantic
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Communications
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.10
Requires-Dist: mcp[cli]>=1.0.0
Requires-Dist: zstandard>=0.20.0
Description-Content-Type: text/markdown

<!-- mcp-name: io.github.Octid-io/osmp -->
# OSMP MCP Server

MCP server for the [Octid Semantic Mesh Protocol (OSMP)](https://octid.io). Gives any MCP-compatible AI client native OSMP capability: encode, decode, and resolve agentic instructions by table lookup. No inference at decode.

**86.8% byte reduction vs JSON. 70.5% vs compiled protobuf. 76.0% token reduction.**

Measured on 29 real-world vectors from MCP, OpenAI, Google A2A, CrewAI, and AutoGen. Compiled .proto verification with protoc 3.21.12.

## Install

```bash
pip install osmp-mcp
```

## Tools

| Tool | What It Does |
|------|-------------|
| `osmp_encode` | Encode structured fields into a SAL instruction |
| `osmp_decode` | Decode SAL to structured fields (handles compound instructions) |
| `osmp_compound_decode` | DAG topology analysis for compound multi-frame instructions |
| `osmp_lookup` | Search the 341-opcode dictionary by namespace and/or keyword |
| `osmp_resolve` | Resolve a domain code (ICD-10-CM, ISO 20022, MITRE ATT&CK) |
| `osmp_batch_resolve` | Resolve multiple domain codes in one call |
| `osmp_discover` | Keyword search across domain corpora |
| `osmp_benchmark` | Run the 55-vector conformance benchmark |

## Resources

| Resource | Contents |
|----------|----------|
| `osmp://system_prompt` | Agent system prompt with grammar and usage rules |
| `osmp://about` | Protocol summary |
| `osmp://dictionary` | Full 341-opcode ASD |
| `osmp://grammar` | SAL formal grammar |
| `osmp://corpora` | Available MDR corpora and stats |
| `osmp://examples` | Worked examples across domains |

## Bundled Corpora

| Corpus | Entries | Binary Size |
|--------|---------|-------------|
| ICD-10-CM FY2026 | 74,719 clinical codes | 477 KB |
| ISO 20022 | 66,956 financial definitions | 1.21 MB |
| MITRE ATT&CK v18.1 | 1,661 entries | 20 KB |

## Links

- [GitHub](https://github.com/octid-io/cloudless-sky)
- [Whitepaper](https://github.com/octid-io/cloudless-sky/blob/main/docs/SAL-efficiency-analysis.md)
- [Benchmark](https://github.com/octid-io/cloudless-sky/tree/main/benchmarks/sal-vs-json)
- [octid.io](https://octid.io)

Apache 2.0. Patent pending (Application #64/007,684).
