mubit-sdk 0.5.0

Public unified Mubit SDK with one Client facade

mubit-sdk

Canonical Python SDK for MuBit.

Install

pip install mubit-sdk

Quickstart

import os

from mubit import Client

client = Client(
    transport=os.getenv("MUBIT_TRANSPORT", "auto"),
    run_id="sdk-python-demo",
    api_key=os.environ["MUBIT_API_KEY"],
)

client.remember(
    session_id="sdk-python-demo",
    agent_id="sdk-quickstart",
    content="If the replay queue stalls, checkpoint before replaying recovery.",
    intent="lesson",
    lesson_type="success",
    lesson_scope="session",
    metadata={"source": "readme"},
)

answer = client.recall(
    session_id="sdk-python-demo",
    query="What should I do before replaying recovery?",
    entry_types=["lesson", "rule"],
)
print(answer.get("final_answer"))

Surface Model

• Low-level raw APIs remain available on client.auth.*, client.core.*, and client.control.*.
• High-level helper APIs live on Client:
  • remember
  • recall
  • get_context
  • archive
  • archive_block
  • dereference
  • memory_health
  • diagnose
  • reflect
  • forget
  • checkpoint
  • register_agent
  • list_agents
  • record_outcome
  • surface_strategies
  • handoff
  • feedback
• Helper APIs use session_id as the ergonomic alias for run_id.

Auto-Capture

The Python package also ships mubit.auto for zero-friction trace capture in MAS learning loops:

from mubit.auto import instrument, observe

instrument()

@observe(name="repair-attempt")
def run_attempt():
    ...

MAS / Learning Loop Example

client.register_agent(
    session_id="sdk-python-demo",
    agent_id="planner",
    role="planner",
    read_scopes=["rule", "lesson", "fact"],
    write_scopes=["lesson", "trace"],
    shared_memory_lanes=["knowledge", "history"],
)

client.checkpoint(
    session_id="sdk-python-demo",
    label="pre-compaction-1",
    context_snapshot="Planner narrowed the failure to token refresh ordering.",
    metadata={"window": 1},
)

strategies = client.surface_strategies(
    session_id="sdk-python-demo",
    lesson_types=["success", "failure"],
    max_strategies=3,
)
print(len(strategies.get("strategies", [])))

Exact References

archived = client.archive(
    session_id="sdk-python-demo",
    artifact_kind="patch_fragment",
    content="--- a/query.py\n+++ b/query.py\n@@ ...",
    labels=["django", "retry"],
    family="patch-repair",
)

exact = client.dereference(
    session_id="sdk-python-demo",
    reference_id=archived["reference_id"],
)
print(exact.get("evidence", {}).get("content"))

Endpoint Resolution

transport defaults to auto (gRPC primary, HTTP fallback).

Endpoint resolution order:

• explicit endpoint / http_endpoint / grpc_endpoint constructor args
• env vars MUBIT_ENDPOINT, MUBIT_HTTP_ENDPOINT, MUBIT_GRPC_ENDPOINT
• shared defaults https://api.mubit.ai and grpc.api.mubit.ai:443

Package Example Lanes

Public adoption scenarios are the primary learning path:

PYTHONPATH=sdk/python/mubit-sdk/src python3 sdk/python/mubit-sdk/examples/public/run_public_examples.py --list
PYTHONPATH=sdk/python/mubit-sdk/src python3 sdk/python/mubit-sdk/examples/public/run_public_examples.py --scenario 01_remember_recall
PYTHONPATH=sdk/python/mubit-sdk/src python3 sdk/python/mubit-sdk/examples/public/run_public_examples.py --scenario 12_auto_capture_observed_loop

Internal raw-smoke scenarios remain available for compatibility and wire-level verification:

PYTHONPATH=sdk/python/mubit-sdk/src python3 sdk/python/mubit-sdk/examples/internal/run_internal_examples.py --list
PYTHONPATH=sdk/python/mubit-sdk/src python3 sdk/python/mubit-sdk/examples/internal/run_internal_examples.py --scenario auth_lifecycle