Python SDK

Provider-agnostic Python client wrapping all MCP tools via JSON-RPC over HTTP.

Install

pip install flux7-memory

PyPI package name

The package is flux7-memory, not mem7. pip install mem7 installs a different, unrelated package.

Or from source :

pip install ./sdk/python

Quick start

from mem7 import Mem7

m = Mem7("http://localhost:9070", token="my-token")

# Store a memory
m.store("deploy.decision", "approved by ops lead, prod deploy greenlit",
        tags=["decision", "deploy"], agent="supervisor")

# Search (returns formatted text)
print(m.search("deployment approval", limit=5))

# Context (returns structured Memory objects)
for mem in m.context("deployment approval", limit=5):
    print(f"{mem.key}: {mem.value}")

API

`Mem7(url, token=None)`

Create a client connected to a running mem7 serve instance.

`store(key, value, tags=None, agent=None, ttl=0)`

Upsert a memory entry.

m.store("user.prefs", "prefers dark mode", tags=["user", "ui"])

`search(query, limit=10, mode="raw", tags=None, agent=None)`

Full-text search, returns formatted markdown text.

results = m.search("dark mode", limit=5, mode="natural")

`context(query, limit=10, **kwargs)`

Same as search but returns a list of Memory objects with structured fields :

for mem in m.context("deploy*", limit=10):
    print(mem.key)       # "deploy.decision"
    print(mem.value)     # "approved by ops lead"
    print(mem.tags)      # ["decision", "deploy"]
    print(mem.agent)     # "supervisor"
    print(mem.updated)   # "2026-05-09T10:30:00Z"

`context_block(query, limit=10, **kwargs)`

Returns a pre-formatted text block for LLM prompt injection :

block = m.context_block("user preferences", limit=10)
# Inject into system prompt or context window

`recall(key=None, tags=None, agent=None, limit=10)`

Recall by key, tags, or agent. Bumps access tracking.

m.recall(key="deploy.decision")
m.recall(tags=["decision"], limit=5)

`list(tags=None, agent=None)`

List keys with metadata (without values).

entries = m.list(tags=["decision"])

`get(path, from_line=None, to_line=None)`

Read a workspace file.

content = m.get("memory/2026-05-09.md")

`forget(key=None, tags=None, agent=None)`

Delete by key and/or tags. Appends a tombstone to the markdown workspace.

m.forget(key="user.prefs")
m.forget(tags=["temp"])

Usage in multi-agent systems

from mem7 import Mem7

m = Mem7("http://localhost:9070", token="shared-token")

# Research agent stores findings
m.store("research.competitors", "Found 3 competitors with similar features",
        tags=["research"], agent="researcher")

# Execution agent checks decisions before acting
decisions = m.context("deployment approved", tags=["decision"], limit=5)
if any("approved" in d.value for d in decisions):
    proceed_with_deploy()

# Supervisor stores policy decisions
m.store("policy.deploy", "requires 2 approvals before prod deploy",
        tags=["policy", "decision"], agent="supervisor")