MAMA

Name: MAMA
Author: jungjaehoon-lifegamez

12stars4forks

How to install MAMA?

jungjaehoon-lifegamez/mama

$git clone https://github.com/jungjaehoon-lifegamez/MAMA

Installs into the current project.

Summary

Local-first operating memory for AI coding agents - Claude Code, Codex, and any MCP client. Tracks decisions and their evolution. SQLite + on-device embeddings, no cloud.

Files · 1

View on GitHub

README.md

1	# MAMA OS - Local Operating Memory for AI Agents
2
3	[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
4	[![Node Version](https://img.shields.io/badge/node-%3E%3D22-brightgreen)](https://nodejs.org)
5	[![LongMemEval 100Q](https://img.shields.io/badge/LongMemEval%20100Q-93%25-blue)](packages/memorybench/)
6	[![Tests](https://img.shields.io/badge/tests-3000%2B%20passing-success)](https://github.com/jungjaehoon-lifegamez/MAMA)
7
8	> Local operating memory that lets AI agents read the board, cite the evidence, and stay inside
9	> explicit boundaries.
10
11	MAMA OS connects chats, docs, decisions, and work logs into a local memory substrate. Agents can use
12	it to search raw evidence, follow relationships, inspect timelines, and produce better briefings
13	instead of guessing from a short prompt.
14
15	This release ships the foundation: raw search/window APIs, graph/entity/timeline APIs, situation
16	packets, trusted provenance, model/tool traces, strict search diagnostics, runtime envelopes, and
17	Context Compile V0. `context_compile` turns those pieces into one selected/rejected/missing
18	evidence packet for a task, and `mama_save` can attach that packet through trusted
19	`context_packet_id` provenance.
20
21	## The Operator Runtime
22
23	The living center of MAMA OS is the trigger loop: an agent that authors its own triggers from
24	recurring situations in your channels, fires them on future messages to recall the right memory,
25	and folds everything into owner situation reports. The loop evolves itself — a review pass retires
26	noisy triggers, and delivered reports that cite a fired trigger feed a success signal back into
27	that trigger's stats.
28
29	What runs continuously today:
30
31	- Trigger loop — agent-authored triggers (keywords + memory query + procedure), deterministic
32	fire → recall → report, near-duplicate authoring gate, citation-based success circuit.
33	- Operator board at `/ui` — a React viewer with four agent-published slots (briefing, action
34	required, decisions, pipeline) rendered live over SSE, a Triggers tab showing the loop's own
35	library, and a native Tasks surface backed by the task ledger. Tasks display workflow status and
36	temporal state separately, so an overdue item is never silently treated as blocked or complete.
37	- Memory promotion — every 6 hours a curation pass promotes durable judgments (pricing rules,
38	standing client preferences, process rules) from recent channel data into decisions. Task states
39	never become memories; the board owns those.
40	- Wiki compilation — promoted decisions chain into an Obsidian wiki organized as an
41	append-only daily journal (`daily/YYYY-MM-DD.md`) plus lesson pages
42	(`lessons/clients\|process\|system`) that strengthen with evidence and get superseded, never
43	deleted.
44	- Hourly self-audit — a conductor pass checks process health, databases, config, and security
45	posture, deduplicating alerts against a state file so the owner hears about a finding once, not
46	every hour.
47
48	## Target Workflow
49
50	MAMA is being built toward a local memory twin that agents can inspect, cite, and act on inside
51	explicit permission boundaries.
52
53	Ask:
54
55	> "Is Project A at risk right now?"
56
57	A search tool finds messages containing "Project A." A MAMA-backed agent reads the board:
58
59	1. The customer said the schedule was fine in email.
60	2. The internal owner changed twice in Slack.
61	3. The core PR is still waiting for review.
62	4. The QA checklist is not closed.
63	5. The same customer changed demo scope at the last minute last month.
64
65	A mature MAMA-backed agent should be able to report:
66
67	- Judgment: schedule risk is high.
68	- Evidence: demo request, review-blocked PR, owner changes, unfinished QA, prior scope-change
69	pattern.
70	- Inference: the customer has not complained yet, but delivery risk is accumulating before the
71	demo.
72	- Missing context: demo scope is not confirmed.
73	- Risk forecast: if review and demo scope do not close today, Friday may turn into a
74	renegotiation.
75	- Next move: assign a PR reviewer, confirm demo scope with the customer, shrink QA to the
76	release-critical path.
77	- Permission boundary: external sending is not allowed, so the agent drafts the message and
78	records the report instead of contacting the customer.
79
80	That is the product direction: not another search box, but the substrate for an extra
81	analyst-operator that can read the company record, separate evidence from inference, forecast the
82	next risk, and only act inside the scope it was given.
83
84	## Why This Matters
85
86	Agents are useful when they can simulate. AlphaGo read the board before choosing the next move. Work
87	agents need the same thing: enough context to reconstruct what happened, infer what may matter, and
88	compare possible next actions.
89
90	Most agents never see that board. They see a prompt, a few files, or one search result. MAMA's job is
91	to make the board visible.
92
93	## North Star
94
95	MAMA OS is moving toward a company memory twin: an append-only substrate of raw recor

People who liked this also viewed

6 picks

Type

excel-mcpA Model Context Protocol (MCP) server for working with Excel (.xlsx) files without requiring Microsoft Excel. Enables AI assistants to create, read, modify, and format spreadsheets using openpyxl, fully in Python and cross-platform.MCP ServersOct 2025570k5
memories-with-lessons-mcp-serverI enhance the existing memory mcp server from the official mcp github, so big thanks and credits for creating thisMCP ServersFeb 2025450k52
anytype-mcpAn MCP server enabling AI assistants to interact with Anytype - your encrypted, local and collaborative wiki - to organize objects, lists, and more through natural language.MCP ServersJul 202669k484
mcp-server-kubernetesMCP Server for kubernetes management commandsMCP ServersJul 202664k1.5k
thoughtboxThoughtbox is an intention ledger for agents. Evaluate AI's decisions against its decision-making.MCP ServersJul 202640k65
google-drive-mcpA Model Context Protocol (MCP) server that provides secure integration with Google Drive, Docs, Sheets, Slides and Calendar. It allows Claude Desktop and other MCP clients to manage files in Google Drive through a standardized interface.MCP ServersJul 202637k189

MAMA

12stars4forks

View on GitHub

How to install MAMA?

jungjaehoon-lifegamez/mama

$git clone https://github.com/jungjaehoon-lifegamez/MAMA

Installs into the current project.

Summary

Local-first operating memory for AI coding agents - Claude Code, Codex, and any MCP client. Tracks decisions and their evolution. SQLite + on-device embeddings, no cloud.

Files · 1

View on GitHub

README.md

1	# MAMA OS - Local Operating Memory for AI Agents
2
3	[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
4	[![Node Version](https://img.shields.io/badge/node-%3E%3D22-brightgreen)](https://nodejs.org)
5	[![LongMemEval 100Q](https://img.shields.io/badge/LongMemEval%20100Q-93%25-blue)](packages/memorybench/)
6	[![Tests](https://img.shields.io/badge/tests-3000%2B%20passing-success)](https://github.com/jungjaehoon-lifegamez/MAMA)
7
8	> Local operating memory that lets AI agents read the board, cite the evidence, and stay inside
9	> explicit boundaries.
10
11	MAMA OS connects chats, docs, decisions, and work logs into a local memory substrate. Agents can use
12	it to search raw evidence, follow relationships, inspect timelines, and produce better briefings
13	instead of guessing from a short prompt.
14
15	This release ships the foundation: raw search/window APIs, graph/entity/timeline APIs, situation
16	packets, trusted provenance, model/tool traces, strict search diagnostics, runtime envelopes, and
17	Context Compile V0. `context_compile` turns those pieces into one selected/rejected/missing
18	evidence packet for a task, and `mama_save` can attach that packet through trusted
19	`context_packet_id` provenance.
20
21	## The Operator Runtime
22
23	The living center of MAMA OS is the trigger loop: an agent that authors its own triggers from
24	recurring situations in your channels, fires them on future messages to recall the right memory,
25	and folds everything into owner situation reports. The loop evolves itself — a review pass retires
26	noisy triggers, and delivered reports that cite a fired trigger feed a success signal back into
27	that trigger's stats.
28
29	What runs continuously today:
30
31	- Trigger loop — agent-authored triggers (keywords + memory query + procedure), deterministic
32	fire → recall → report, near-duplicate authoring gate, citation-based success circuit.
33	- Operator board at `/ui` — a React viewer with four agent-published slots (briefing, action
34	required, decisions, pipeline) rendered live over SSE, a Triggers tab showing the loop's own
35	library, and a native Tasks surface backed by the task ledger. Tasks display workflow status and
36	temporal state separately, so an overdue item is never silently treated as blocked or complete.
37	- Memory promotion — every 6 hours a curation pass promotes durable judgments (pricing rules,
38	standing client preferences, process rules) from recent channel data into decisions. Task states
39	never become memories; the board owns those.
40	- Wiki compilation — promoted decisions chain into an Obsidian wiki organized as an
41	append-only daily journal (`daily/YYYY-MM-DD.md`) plus lesson pages
42	(`lessons/clients\|process\|system`) that strengthen with evidence and get superseded, never
43	deleted.
44	- Hourly self-audit — a conductor pass checks process health, databases, config, and security
45	posture, deduplicating alerts against a state file so the owner hears about a finding once, not
46	every hour.
47
48	## Target Workflow
49
50	MAMA is being built toward a local memory twin that agents can inspect, cite, and act on inside
51	explicit permission boundaries.
52
53	Ask:
54
55	> "Is Project A at risk right now?"
56
57	A search tool finds messages containing "Project A." A MAMA-backed agent reads the board:
58
59	1. The customer said the schedule was fine in email.
60	2. The internal owner changed twice in Slack.
61	3. The core PR is still waiting for review.
62	4. The QA checklist is not closed.
63	5. The same customer changed demo scope at the last minute last month.
64
65	A mature MAMA-backed agent should be able to report:
66
67	- Judgment: schedule risk is high.
68	- Evidence: demo request, review-blocked PR, owner changes, unfinished QA, prior scope-change
69	pattern.
70	- Inference: the customer has not complained yet, but delivery risk is accumulating before the
71	demo.
72	- Missing context: demo scope is not confirmed.
73	- Risk forecast: if review and demo scope do not close today, Friday may turn into a
74	renegotiation.
75	- Next move: assign a PR reviewer, confirm demo scope with the customer, shrink QA to the
76	release-critical path.
77	- Permission boundary: external sending is not allowed, so the agent drafts the message and
78	records the report instead of contacting the customer.
79
80	That is the product direction: not another search box, but the substrate for an extra
81	analyst-operator that can read the company record, separate evidence from inference, forecast the
82	next risk, and only act inside the scope it was given.
83
84	## Why This Matters
85
86	Agents are useful when they can simulate. AlphaGo read the board before choosing the next move. Work
87	agents need the same thing: enough context to reconstruct what happened, infer what may matter, and
88	compare possible next actions.
89
90	Most agents never see that board. They see a prompt, a few files, or one search result. MAMA's job is
91	to make the board visible.
92
93	## North Star
94
95	MAMA OS is moving toward a company memory twin: an append-only substrate of raw recor

People who liked this also viewed

6 picks

Type