Skip to content
Back to search

memorizz

MemoRizz: A Python library serving as a memory layer for AI applications. Leverages popular databases and storage solutions to optimize memory usage. Provides utility classes and methods for efficient data management.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

It has changed since its scan on 11 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/richmondalake/memorizz with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/richmondalake/memorizz'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/richmondalake--memorizz/badge.svg)](/listings/richmondalake--memorizz)

HTML

<a href="/listings/richmondalake--memorizz"><img src="/api/listings/richmondalake--memorizz/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Requires Python 3.10+. Install and launch the CLI with a cloud model:

python -m pip install memorizz
export OPENAI_API_KEY="your-key"
memorizz

For a fully local setup, use [Ollama](docs/getting-started/cli.md#option-a-local-no-api-key-ollama). The [installation guide](docs/getting-started/installation.md) covers `uv`, `pipx`, npm, Homebrew, and optional integrations.

Save credentials without putting them in shell history, and choose memory storage independently from the model:

memorizz config set OPENAI_API_KEY  # hidden prompt
memorizz memory configure          # guided memory-provider setup
memorizz notion connect            # Notion library + separate vector provider
memorizz config path               # actual save target and precedence

These commands save defaults for the next launch; they do not switch a running agent or migrate memory. See [CLI configuration](docs/getting-started/cli.md#configuration).

Build a memory-first agent in Python using the same environment:

from memorizz import MemAgentBuilder

agent = (
    MemAgentBuilder()
    .with_instruction("Remember useful context and use it to help the user.")
    .with_llm_config({"provider": "openai", "model": "gpt-4o-mini"})
    .with_memory_ids("my-assistant")
    .build_and_save()
)

scope = {"memory_id": "my-assistant", "user_id": "alice", "thread_id": "intro"}
with agent.lifecycle(close_memory_provider=True):
    agent.run("I work on payments systems.", **scope)
    print(agent.run("What did I tell you about my work?", **scope))

Memory persists under `~/.memorizz/memory` by default. Save `agent.agent_id` to [restore the agent later](docs/getting-started/python-sdk-quickstart.md#5-restore-it-later).

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 18 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. {
  2. "name": "memorizz",
  3. "owner": {
  4. "name": "Richmond Alake",
  5. "url": "https://github.com/RichmondAlake"
  6. },
  7. "metadata": {
  8. "description": "MemoRizz plugins: durable project memory for coding agents."
  9. },
  10. "plugins": [
  11. {
  12. "name": "memorizz",
  13. "source": "./plugins/claude-code/memorizz",
  14. "description": "Durable, searchable project memory for Claude Code, shared with Codex and your MemoRizz agents.",
  15. "category": "productivity"
  16. }
  17. ]
  18. }
README.md · README · 230 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. # Memorizz
  2. **The intelligence plane for memory-first agents.**
  3. Memorizz is a Python framework that connects persistent memory, models, prompts,
  4. and reasoning so agents can recall what matters, reuse experience, and make
  5. better decisions across sessions.
  6. [Documentation](https://richmondalake.github.io/memorizz/) ·
  7. [PyPI](https://pypi.org/project/memorizz/) ·
  8. [Examples](examples/zero_to_hero/README.md) ·
  9. [Changelog](CHANGELOG.md)
  10. <picture>
  11. <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/RichmondAlake/memorizz/main/docs/assets/screenshots/dashboard-dark.png">
  12. <img alt="The MemoRizz agent dashboard: runs, success rate, run time, model calls, tokens and spend across every agent, a runs-over-time chart, and what needs attention." src="https://raw.githubusercontent.com/RichmondAlake/memorizz/main/docs/assets/screenshots/dashboard-light.png">
  13. </picture>
  14. *The agent dashboard in the local UI (`memorizz ui`): runs, success rate, run
  15. time, model calls, tokens and spend across every agent, and what needs
  16. attention.*
  17. ## Memory management
  18. Memory is the foundation: retain information, retrieve relevant context, and
  19. refine what an agent knows over time.
  20. | Memory | What it gives your agents |
  21. |---|---|
  22. | [Episodic](docs/memory-types/episodic.md) | Conversation history, past interactions, and summaries. |
  23. | [Semantic](docs/memory-types/semantic.md) | Knowledge bases, entity facts, preferences, and personas. |
  24. | [Procedural](docs/memory-types/procedural.md) | Registered tools, recorded workflows, and reusable skills. |
  25. | [Short-term](docs/memory-types/short-term.md) | Working context and semantic caching. |
  26. | [Shared](docs/memory-types/shared.md) | A shared workspace for agent coordination and handoffs. |
  27. - **Persist and isolate:** scope memory by agent, user, and conversation. Start
  28. with local filesystem storage, or use MongoDB, Oracle, or
  29. [Notion with separate vector storage](docs/memory-providers/notion.md) through the same
  30. provider contract.
  31. - **Ingest and retrieve:** turn files and folders into a knowledge base with
  32. configurable chunking, embeddings, and semantic or hybrid search.
  33. - **Manage context:** deduplicate retrieved memories, summarize and compact
  34. history, inspect recall decisions, and control retention and forgetting.
  35. - **Forget the way Generative Agents do:** retrieval ranks memories by recency
  36. since last use, stored importance and relevance, recalls reinforce a memory,
  37. and a governed planner suppresses, never deletes, records whose retention
  38. decayed; every suppression is approved and reversible. See the
  39. [forgetting mechanism](docs/guides/forgetting-mechanism.md).
  40. Explore [memory concepts and scope](docs/getting-started/concepts.md),
  41. [storage providers](docs/memory-providers/filesystem.md), and
  42. [context efficiency](docs/guides/context-efficiency.md).
  43. ## Agent framework
  44. Build persistent agents with `MemAgent` and `MemAgentBuilder`. Choose
  45. `assistant`, `workflow`, or `deep_research` modes and connect cloud or local
  46. [model providers](docs/getting-started/model-providers.md).
  47. - **Tools and integrations:** Python functions, MCP servers, internet search,
  48. governed browser control, [desktop control](docs/desktop-control/index.md)
  49. of Mac apps (computer use), sandboxes, and external skills. MCP tools are
  50. stored in toolbox memory and found by meaning. A
  51. [tool call cache](docs/guides/context-efficiency.md#tool-call-cache) reuses
  52. the results of repeated calls to tools that are safe to reuse.
  53. - **Persona:** give an agent a name, role, goals and background. It changes
  54. when someone asks, and with
  55. [persona learning](docs/guides/persona-learning.md) on, it can grow in the
  56. background from what the agent is used for, with every change quoted and
  57. undoable.
  58. - **Coordination:** delegate work to specialist agents, share memory, run
  59. scheduled automations, and start [timers](docs/guides/automations.md#timers)
  60. that count down live in the CLI and the chat.
  61. - **Interfaces:** use the Python SDK, interactive CLI, local web UI, or expose
  62. Memorizz as an MCP server.
  63. Start with the [SDK quickstart](docs/getting-started/python-sdk-quickstart.md)
  64. or the [CLI guide](docs/getting-started/cli.md).
  65. ## Harness
  66. `MetaHarness` runs tasks through Codex, Claude Code, OpenHands, DeepSeek, pi,
  67. Hermes, or native MemAgent workers, and a MemAgent can hand parts of a request
  68. to delegates that run on those harnesses. It carries memory scope and learning evidence across
  69. harnesses while managing permissions, approvals, workspace isolation, budgets,
  70. cancellation, and host-side verification.
  71. Use an external harness for an entire task or let a MemAgent call one as a
  72. specialist tool. Inspect installed adapters with `memorizz harness doctor`.
  73. See the [MetaHarness guide](docs/guides/meta-harness.md) and
  74. [hands-on examples](examples/metaharness/README.md).
  75. <picture>
  76. <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/RichmondAlake/memorizz/main/docs/assets/screenshots/harnesses-dark.png">
  77. <img alt="The MemoRizz harness dashboard: launch a task on a harness, see which harnesses are ready on this machine, and follow plans and comparisons across Codex, Claude Code and MemAgents." src="https://raw.githubusercontent.com/RichmondAlake/memorizz/main/docs/assets/screenshots/harnesses-light.png">
  78. </picture>
  79. *The harness dashboard: run a memory-grounded task on one harness, chain
  80. harnesses into a plan, or compare them side by side, with approvals and
  81. verification.*
  82. ### Memory for Codex and Claude Code
  83. MemoRizz is also a plugin for Codex and Claude Code. The agent starts each
  84. session with the project's facts and the last session's summary; each turn and
  85. a summary of each session are saved as episodic memory; and the agent can
  86. search past sessions, save and correct facts, record entities and load project
  87. documents. Both agents share the memory, locally or on a hosted MemoRizz MCP
  88. server for a team ([deploy/mcp-server](deploy/mcp-server/README.md)). Each
  89. session also appears on the UI's Harnesses page with its full trajectory, so
  90. you can follow a Codex session and a Claude Code session step by step and
  91. compare them.
  92. ```bash
  93. memorizz plugin install codex # or: memorizz plugin install claude-code
  94. ```
  95. Or `codex plugin marketplace add RichmondAlake/memorizz` /
  96. `claude plugin marketplace add RichmondAlake/memorizz`, then install
  97. `memorizz@memorizz`. See [MemoRizz in Codex and Claude Code](docs/guides/coding-agent-plugins.md),
  98. and [examples/coding_agent_plugins](examples/coding_agent_plugins/README.md) for
  99. Codex and Claude Code sharing one project's memory.
  100. ## Continual learning and the intelligence plane
  101. The **intelligence plane** is the model layer and everything that helps it
  102. think: the LLM, prompts, reasoning strategies such as planning, reflection, and
  103. tool selection, plus the memory and retrieval that supply context. Its job is
  104. **decision quality**. A bad plan or a hallucinated tool call is an intelligence
  105. plane problem.
  106. Memorizz makes memory the foundation of this plane. Its opt-in continual
  107. learning loop captures tool workflows and their outcomes, distills repeated
  108. successful procedures into reusable skills, retrieves relevant active skills,
  109. and monitors their performance. Shadow evaluation and review support skill
  110. activation; skills that stop working can be demoted.
  111. The learning control plane adds traceable outcome evidence, bounded retrieval,
  112. and reversible forgetting. Together, these capabilities help agents retain,
  113. recall, reuse, and refine experience.
  114. Memorizz also fine-tunes the small models behind memory on your own
  115. memories: decision models that file each memory in the right store,
  116. embedding models (EmbeddingGemma 2, Qwen3-Embedding) and rerankers. The
  117. datasets stay in the connected memory provider, each model is scored against
  118. the one in use on held-out examples, and you swap it in (or back) from the
  119. UI, the CLI or MCP.
  120. Where memory used to decide with a similarity score and a fixed threshold, an
  121. agent can ask a [decision model](docs/guides/decision-models.md) instead:
  122. Liquid's d1 on your machine, TypeSafe's Jev, OpenAI Decisions, or one
  123. fine-tuned in Memorizz. It answers whether a message holds anything worth
  124. keeping, which memory a request needs, whether a cached answer fits the
  125. question as asked, how a new note relates to stored ones, and which tool a
  126. request needs, if any.
  127. Read about [continual learning](docs/guides/continual-learning.md),
  128. [persona learning](docs/guides/persona-learning.md),
  129. [fine-tuning](docs/guides/fine-tuning.md),
  130. [decision models](docs/guides/decision-models.md), the
  131. [learning control plane](docs/guides/learning-control-plane.md), the
  132. [forgetting mechanism](docs/guides/forgetting-mechanism.md), and
  133. [evaluation methods](docs/evaluation-suite.md).
  134. ## Quick start
  135. Requires Python 3.10+. Install and launch the CLI with a cloud model:
  136. ```bash
  137. python -m pip install memorizz
  138. export OPENAI_API_KEY="your-key"
  139. memorizz
  140. ```
  141. For a fully local setup, use [Ollama](docs/getting-started/cli.md#option-a-local-no-api-key-ollama).
  142. The [installation guide](docs/getting-started/installation.md) covers `uv`,
  143. `pipx`, npm, Homebrew, and optional integrations.
  144. Save credentials without putting them in shell history, and choose memory
  145. storage independently from the model:
  146. ```bash
  147. memorizz config set OPENAI_API_KEY # hidden prompt
  148. memorizz memory configure # guided memory-provider setup
  149. memorizz notion connect # Notion library + separate vector provider
  150. memorizz config path # actual save target and precedence
  151. ```
  152. These commands save defaults for the next launch; they do not switch a running
  153. agent or migrate memory. See [CLI configuration](docs/getting-started/cli.md#configuration).
  154. Build a memory-first agent in Python using the same environment:
  155. ```python
  156. from memorizz import MemAgentBuilder
  157. agent = (
  158. MemAgentBuilder()
  159. .with_instruction("Remember useful context and use it to help the user.")
  160. .with_llm_config({"provider": "openai", "model": "gpt-4o-mini"})
  161. .with_memory_ids("my-assistant")
  162. .build_and_save()
  163. )
  164. scope = {"memory_id": "my-assistant", "user_id": "alice", "thread_id": "intro"}
  165. with agent.lifecycle(close_memory_provider=True):
  166. agent.run("I work on payments systems.", **scope)
  167. print(agent.run("What did I tell you about my work?", **scope))
  168. ```
  169. Memory persists under `~/.memorizz/memory` by default. Save `agent.agent_id` to
  170. [restore the agent later](docs/getting-started/python-sdk-quickstart.md#5-restore-it-later).
  171. ## Explore
  172. - [Memory-first tutorials](examples/zero_to_hero/README.md)
  173. - [Persistent notes and searchable history: filesystem and Oracle notebooks](examples/context_engineering/README.md)
  174. - [MCP connectivity](docs/guides/mcp-connectivity.md) and [MCP server](docs/guides/mcp-server.md)
  175. - [Automations and timers](docs/guides/automations.md) and [multi-user applications](docs/guides/multi-tenant.md)
  176. - [Observability](docs/observability-ui.md) and [evaluation suite](docs/evaluation-suite.md)
  177. - [Usage, cost and memory analytics](docs/usage-analytics.md)
  178. - [Persona learning](docs/guides/persona-learning.md) and [account-owned persona evolution](docs/persona-evolution-hosts.md)
  179. - [API reference](docs/reference/python-api.md), [configuration](docs/reference/configuration.md), and [troubleshooting](docs/troubleshooting.md)
  180. ## License
  181. Source available under [PolyForm Noncommercial 1.0.0](LICENSE). See [NOTICE](NOTICE).
  182. Memorizz is experimental: APIs may change and it has not undergone security
  183. hardening for production workloads.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.