OSMP MCP Server — Agentic AI instruction encoding. 86.8% byte reduction vs JSON. Inference-free decode. Any channel.
Project description
OSMP MCP Server
MCP server for the Octid Semantic Mesh Protocol (OSMP). Gives any MCP-compatible AI client native OSMP capability: encode, decode, validate, and resolve agentic instructions by table lookup. No inference at decode.
Install
pip install osmp-mcp
Or with uvx:
uvx osmp-mcp
What it does
- osmp_encode — Encode structured fields into a SAL instruction
- osmp_decode — Decode SAL to structured fields
- osmp_validate — Check composed SAL against composition rules before emission
- osmp_lookup — Search the opcode dictionary by namespace/keyword
- osmp_resolve — Resolve a domain code from D:PACK/BLK (ICD-10, ISO 20022, MITRE ATT&CK)
- osmp_discover — Search domain corpora by keyword
- osmp_benchmark — Run the canonical conformance suite
- osmp_compound_decode — Analyze DAG topology and loss tolerance
- osmp_batch_resolve — Multiple domain codes in one call
342 opcodes. 26 namespaces. Three MDR corpora (124,215 resolvable codes).
Composition Rules
SAL has a grammar (how to write a legal instruction) and a usage doctrine (which instruction to write). Both matter.
The taco test: K:ORD is financial order entry (ISO 20022). "Order me some tacos" is NL passthrough, not K:ORD[TACOS]. Read the ASD definition, not the mnemonic.
The lookup gate: Always call osmp_lookup before composing. If the opcode doesn't exist in the dictionary, it doesn't exist in the protocol.
Validation: Call osmp_validate after composing and before emitting. It catches hallucinated opcodes, missing consequence classes, namespace-as-target errors, byte inflation, and other structural violations.
Full composition doctrine: SAL-usage-doctrine-v1.md
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file osmp_mcp-1.0.22.tar.gz.
File metadata
- Download URL: osmp_mcp-1.0.22.tar.gz
- Upload date:
- Size: 1.8 MB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
10dfa857914fbb5d189f94734bb2d91d4bb0c9f8741f6e8943acce75c3b79038
|
|
| MD5 |
6d7abcf851f1b143d1ee52b8ee2b91e5
|
|
| BLAKE2b-256 |
734b19aca4b6f06f4d6c42fd4f0253ffee81ead239a8220c7853eaaf544d3bee
|
File details
Details for the file osmp_mcp-1.0.22-py3-none-any.whl.
File metadata
- Download URL: osmp_mcp-1.0.22-py3-none-any.whl
- Upload date:
- Size: 1.8 MB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
118ea17aa18832c443d13b01b54055a4004a02982d8384fc004244ff064cacfd
|
|
| MD5 |
c85556d990ea75f218a8c1dbfaac5a13
|
|
| BLAKE2b-256 |
1f7011fc8392f942888f32545624aed63a38082b4b51b6ad010c289e314baf31
|