AXM  ·  Axiom
axm-genesis v1.2.0  ·  axm-chat v0.1.0  ·  Apache 2.0

Your AI conversations
as sovereign knowledge

Turn Claude, ChatGPT, and Gemini exports into cryptographically signed, offline-queryable knowledge shards. Your data. Your keys. Your machine.

Vendor APIs are memory black holes.

You spend hours reasoning through architecture, strategic planning, and complex decisions with LLMs. That cognitive work is trapped inside fragmented chat interfaces controlled by vendors.

When you need to remember why a specific path was chosen or what alternatives were rejected, the context is gone. Search fails because the decisions were never formally committed. The data is locked in the cloud.

Vector databases are search engines. This is version control for knowledge.

The AXM protocol was built to verify autonomous robotics doctrine offline. axm-chat applies that exact same cryptographic rigidity to your personal knowledge base.

No black boxes. Just files.

A "shard" is not a proprietary database. It is a standard directory. You can zip it, email it, or query it with any data science tool. The security is in the cryptography, not the storage medium.

Click manifest.json to inspect the structure. This is the exact artifact axm-chat produces. axm-core mounts these directories directly into DuckDB at query time.

# A compiled AXM conversation shard
~/.axm/shards/chat_My_Conversation_abc123/

manifest.json           # metadata + Merkle root
content/source.txt       # full transcript
graph/claims.parquet     # extracted structured claims
sig/manifest.sig         # ML-DSA-44 signature
ext/lineage@1.parquet      # supersession chain
BLAKE3 Merkle Trees
Every claim and text block is hashed into a strict Merkle tree. Verify structural integrity of any shard entirely offline — no central server required.
ML-DSA-44 Signatures
Post-quantum signatures (FIPS 204) ensure that once a shard is compiled, the authorship and timestamp are mathematically immutable. Ed25519 also supported.
DuckDB + Parquet
Shards are stored locally as compressed Parquet files. At query time, the Spectra engine mounts them in-memory via DuckDB for instant relational analysis.
Lineage + References
When a decision shard supersedes a conversation shard, lineage@1 records the chain. references@1 links every decision claim back to the exact conversation turns that produced it.
# Install the AXM stack (Python 3.10+)
pip install -e ./axm-genesis    # cryptographic compiler + verifier
pip install -e ./axm-core       # Spectra DuckDB query engine
pip install -e ./axm-chat       # chat spoke

# Export your conversations, then import
axm-chat import ./conversations.json

# Index episodes + extract decisions using a local model
ollama serve && ollama pull mistral
axm-chat distill

# Query your sovereign data
axm-chat query "what decisions have we made"
axm-chat query "what changed since january"
axm-chat verify                # offline Merkle + signature check

All data stays in ~/.axm/shards/ on your machine. Verify any shard entirely offline: axm-verify ./shard_dir