AutoGineer
Generated banner for memmy-agent
Generated logo for memmy-agent

memmy-agent

Agent plugin

First committed by ZongYue99

Published in MemTensor/memmy-agent. Publisher page

2,210 GitHub stars

What it is

Local-first agent memory substrate with desktop and CLI surfaces.
— the description by ZongYue99

Summary from the brain’s metadata

Local-first agent memory substrate with desktop and CLI surfaces.

How it works

Memory model: Not stated

AutoGineer overview, from the brain’s own files

Folders: .github (agent config, 48 files); App (source code, 2536 files); Knowledge (knowledge base, 20 files); Memory (memory, 737 files); docs (docs, 388 files); scripts (scripts, 103 files); tests (tests, 43 files).

Folders
FolderRoleFiles
.githubagent-config48
Appsource-code2536
Knowledgeknowledge-base20
Memorymemory737
docsdocs388
scriptsscripts103
teststests43
Migrationsfolder48
pluginsfolder36

Requirements

Runtimes: Node.js >=20

API keys (by name): OPENAI_API_KEY

Set it up

  1. This brain runs code automatically: 4 scripts. Review Memory/installers/install.ps1, Memory/installers/install.sh, README.md, scripts/install.sh before opening the folder in Claude Code. No step below turns a hook on for you.

    • Memory/installers/install.ps1: pwsh Memory/installers/install.ps1
    • Memory/installers/install.sh: sh Memory/installers/install.sh
    • README.md: curl -fsSL https://raw.githubusercontent.com/MemTensor/memmy-agent/main/scripts/install.sh | bash
    • scripts/install.sh: sh scripts/install.sh
  2. Not scanned by AutoGineerThe author’s README also describes installing it from a clone of its repository. Read its install section: AutoGineer did not scan that path. https://github.com/MemTensor/memmy-agent/blob/e640129a40807522c6c1e4db0c2d3f4f0cb3477d/README.md#how-to-use-memmy

  3. First committed by ZongYue99. Published in MemTensor/memmy-agent. Licence: MIT License.

Usage

AutoGineer’s guidance for this type of brain

Install the plugin in your coding agent, then work as usual: it records and recalls what the agent learns.

Documentation by ZongYue99

A licence file holds personal contact details, so AutoGineer links to the files instead of copying them. Read the documentation at the source

512 files · 512 not served here

Advantages

  • Open licence (MIT) (a rule-derived fact from the brain’s files)

Limitations

  • Needs an API key (OPENAI_API_KEY) (a rule-derived fact from the brain’s files, evidence: README.md)
  • Link-only: AutoGineer links to the files instead of serving them (a rule-derived fact from the brain’s files)

Security

Risk level: Medium risk

This brain runs

Agent instructions that run or fetch something (64)

  • App/memmy-agent/src/skills/agent-memory-onboarding/SKILL.md: Inspect the installed Agent, install the rendered Memmy Skill through its native extension mechanism, and persist one declarative history recipe that the backend can reuse without another Agent session.
  • App/memmy-agent/src/skills/agent-memory-onboarding/SKILL.md: Run it only when the current task explicitly names `$agent-memory-onboarding`.
  • App/memmy-agent/src/skills/agent-memory-onboarding/SKILL.md: `operation`: `connect`, `install`, or `uninstall`
  • App/memmy-agent/src/skills/agent-memory-onboarding/SKILL.md: Before history discovery or any `connect` or `install` write, prove that `agent_name` identifies a product already installed on this machine.
  • App/memmy-agent/src/skills/agent-memory-onboarding/SKILL.md: Run Linux-side inspection with `wsl -d <distribution> -- ...`.
  • App/memmy-agent/src/skills/agent-memory-onboarding/SKILL.md: Render, install, and verify the Memmy Skill.
  • App/memmy-agent/src/skills/agent-memory-onboarding/SKILL.md: Pass the Installation Identity Gate, then install only the verified target Agent's Memmy Skill.
  • App/memmy-agent/src/skills/agent-memory-onboarding/SKILL.md: Discover its native Skill location, render the exact source-specific file, install it, verify it, and call:
  • App/memmy-agent/src/skills/agent-memory-onboarding/SKILL.md: `install` does not claim automatic-scan readiness.
  • App/memmy-agent/src/skills/agent-memory-onboarding/SKILL.md: Run the candidate recipe twice and require stable unique ids plus at least one complete user-to-assistant turn.
  • App/memmy-agent/src/skills/agent-memory-onboarding/SKILL.md: Prefer the Agent's native Skill directory and install the returned file as `memmy-memory/SKILL.md`.
  • App/memmy-agent/src/skills/agent-memory-onboarding/SKILL.md: if the command is outside `PATH`, locate the configured Memmy CLI and run that binary rather than skipping health validation.
  • App/memmy-agent/src/skills/agent-memory-onboarding/SKILL.md: Run the extraction twice and confirm identical, unique message ids.
  • App/memmy-agent/src/skills/agent-memory-onboarding/SKILL.md: For SQLite, run one complete read-only `SELECT` with no semicolon and no `?`, `$name`, or `:name` placeholders.
  • App/memmy-agent/src/skills/computer-history/SKILL.md: **Read events selectively.** A single line can carry a whole accessibility tree and run to tens of thousands of characters.
  • App/memmy-agent/src/skills/computer-use/SKILL.md: browser and desktop tools execute actions.
  • App/memmy-agent/src/skills/computer-use/SKILL.md: Execute directly when the task already provides the required file, save location, credentials, verification code or choice.
  • App/memmy-agent/src/skills/computer-use/SKILL.md: Execute the actions requested by the user, including a final submission or write when it is already part of the task, without requesting the same authorization again.
  • App/memmy-agent/src/skills/docx/SKILL.md: node scripts/render_docx.mjs input.docx --output_dir /path/to/qa
  • App/memmy-agent/src/skills/docx/SKILL.md: Scripts are intended to be run from a writable working directory, while this skill directory remains the source of reusable helpers.
  • App/memmy-agent/src/skills/github/SKILL.md: gh run list --repo owner/repo --limit 10
  • App/memmy-agent/src/skills/github/SKILL.md: View a run and see which steps failed:
  • App/memmy-agent/src/skills/github/SKILL.md: gh run view <run-id> --repo owner/repo
  • App/memmy-agent/src/skills/github/SKILL.md: gh run view <run-id> --repo owner/repo --log-failed
  • App/memmy-agent/src/skills/pdf/SKILL.md: If unavailable, install Poppler or ask the user to review the output locally.
  • App/memmy-agent/src/skills/pdf/SKILL.md: If installation isn't possible in this environment, tell the user which dependency is missing and how to install it locally.
  • App/memmy-agent/src/skills/pptx/SKILL.md: Run `scripts/check_deck.mjs <deck-or-directory> --json`;
  • App/memmy-agent/src/skills/skill-creator/SKILL.md: Run the validator.
  • App/memmy-agent/src/skills/skill-creator/SKILL.md: Commands that should be run exactly.
  • App/memmy-agent/src/skills/skill-creator/SKILL.md: Prefer Python, shell, or other scripts that can run through an explicit interpreter command.
  • App/memmy-agent/src/skills/skill-creator/SKILL.md: Run the bundled quick validator before considering a skill ready:
  • App/memmy-agent/src/skills/skill-creator/SKILL.md: python3 {baseDir}/scripts/quick-validate.py <skill-dir>
  • App/memmy-agent/src/skills/skill-creator/SKILL.md: If validation fails, fix the reported issue and run the command again.
  • App/memmy-agent/src/skills/tmux/SKILL.md: On Windows, use WSL and install tmux inside WSL.
  • App/memmy-agent/src/skills/tmux/SKILL.md: tmux -S "$SOCKET" send-keys -t agent-1 "cd /tmp/project1 && codex --yolo 'Fix bug X'" Enter
  • App/memmy-agent/src/skills/tmux/SKILL.md: tmux -S "$SOCKET" send-keys -t agent-2 "cd /tmp/project2 && codex --yolo 'Fix bug Y'" Enter
  • App/memmy-agent/src/skills/tmux/SKILL.md: `pnpm install` first before running codex in fresh clones
  • App/memmy-agent/src/skills/tmux/SKILL.md: Codex needs `--yolo` or `--full-auto` for non-interactive fixes
  • App/memmy-agent/src/skills/ui-craft/SKILL.md: For independent static HTML whose files run as written without a build, backend, or application route, pass the permitted local `.html`/`.htm` path directly to `browser_navigate`.
  • App/memmy-agent/src/skills/ui-craft/SKILL.md: Run the anti-template review against the screenshots.
  • App/memmy-agent/src/skills/ui-craft/SKILL.md: If browser tools are unavailable, run the strongest static and project checks available and explicitly report that browser validation was not performed.
  • App/memmy-agent/src/skills/update-setup/SKILL.md: Use `exec` to run `memmy --version`.
  • App/memmy-agent/src/skills/update-setup/SKILL.md: Then collect install clues with `exec`.
  • App/memmy-agent/src/skills/update-setup/SKILL.md: Use the clues only to suggest a likely install method.
  • App/memmy-agent/src/skills/update-setup/SKILL.md: The install method must come from a user answer or confirmation, not from inference alone.
  • App/memmy-agent/src/skills/update-setup/SKILL.md: Treat a request like "generate upgrade config", "set up updates", or "use update-setup" as the goal, not as confirmation of the install method.
  • App/memmy-agent/src/skills/update-setup/SKILL.md: Install clues are only hints for the question summary;
  • App/memmy-agent/src/skills/update-setup/SKILL.md: In the initial setup turn, write `skills/update/SKILL.md` only if the user message already contains an explicit install method and proxy answer.
  • App/memmy-agent/src/skills/update-setup/SKILL.md: Hard stop rule: after collecting install clues without an explicit install method from the user, do not call `write_file` for any path ending in `skills/update/SKILL.md` in that same assistant turn.
  • App/memmy-agent/src/skills/update-setup/SKILL.md: **Question 1 — Install method:**
  • App/memmy-agent/src/skills/update-setup/SKILL.md: question: "I found these install clues: <SUMMARY>. Which update method should this workspace use?"
  • App/memmy-agent/src/skills/update-setup/SKILL.md: Determine the upgrade command from the install method:
  • App/memmy-agent/src/skills/update-setup/SKILL.md: | Method | Command | |--------|---------| | npm | `npm install -g memmy-agent@latest` | | pnpm | `pnpm add -g memmy-agent@latest` | | yarn | `yarn global add memmy-agent@latest` | | source | `cd <SOURCE_CHECKOUT> && git pull && npm install && npm run build` | · uses the network
  • App/memmy-agent/src/skills/update-setup/SKILL.md: Determine the preflight check from the install method:
  • App/memmy-agent/src/skills/update-setup/SKILL.md: 2. Use `exec` to run the preflight check: <PREFLIGHT_CHECK>. If it fails, stop and tell the user to rerun `update-setup` because the saved install method no longer matches this environment.
  • App/memmy-agent/src/skills/update-setup/SKILL.md: 3. Use `exec` to run the upgrade command: <UPGRADE_COMMAND>
  • App/memmy-agent/src/skills/update-setup/SKILL.md: 5. Tell the user the new version. Say: "Run `/restart` to restart memmy and apply the update. If `/restart` is unavailable in this channel, restart the memmy process manually."
  • App/memmy-agent/src/skills/weather/SKILL.md: curl -s "wttr.in/London?format=3" · uses the network
  • App/memmy-agent/src/skills/weather/SKILL.md: curl -s "wttr.in/London?format=%l:+%c+%t+%h+%w" · uses the network
  • App/memmy-agent/src/skills/weather/SKILL.md: curl -s "wttr.in/London?T" · uses the network
  • App/memmy-agent/src/skills/weather/SKILL.md: PNG: `curl -s "wttr.in/Berlin.png" -o /tmp/weather.png` · uses the network
  • App/memmy-agent/src/skills/weather/SKILL.md: curl -s "https://api.open-meteo.com/v1/forecast?latitude=51.5&longitude=-0.12&current_weather=true" · uses the network
  • App/memmy-agent/src/skills/xlsx/SKILL.md: After adding or changing formulas run `scripts/recalculate_workbook.mjs --input <file> --output <file> --json`.
  • App/memmy-agent/src/skills/xlsx/SKILL.md: Run `scripts/check_workbook.mjs <file> --json`;

Install scripts (4)

  • Memory/installers/install.ps1: pwsh Memory/installers/install.ps1 · uses the network
  • Memory/installers/install.sh: sh Memory/installers/install.sh · uses the network
  • README.md: curl -fsSL https://raw.githubusercontent.com/MemTensor/memmy-agent/main/scripts/install.sh | bash · uses the network
  • scripts/install.sh: sh scripts/install.sh · uses the network
  • Medium risk: runs-instruction-approval-bypass (App/memmy-agent/src/skills/tmux/SKILL.md:81)
  • Medium risk: runs-instruction-approval-bypass (App/memmy-agent/src/skills/tmux/SKILL.md:82)
  • Medium risk: runs-instruction-approval-bypass (App/memmy-agent/src/skills/tmux/SKILL.md:101)
  • Medium risk: runs-instruction-instruction-network (App/memmy-agent/src/skills/update-setup/SKILL.md:85)
  • Medium risk: runs-instruction-instruction-network (App/memmy-agent/src/skills/weather/SKILL.md:16)
  • Medium risk: runs-instruction-instruction-network (App/memmy-agent/src/skills/weather/SKILL.md:22)
  • Medium risk: runs-instruction-instruction-network (App/memmy-agent/src/skills/weather/SKILL.md:28)
  • Medium risk: runs-instruction-instruction-network (App/memmy-agent/src/skills/weather/SKILL.md:38)
  • Medium risk: runs-instruction-instruction-network (App/memmy-agent/src/skills/weather/SKILL.md:44)
  • Medium risk: runs-install-script-network-variable-host (Memory/installers/install.ps1)
  • Medium risk: network-variable-host (Memory/installers/install.ps1:17)
  • Medium risk: network-variable-host (Memory/installers/install.ps1:18)
  • Medium risk: runs-install-script-network-variable-host (Memory/installers/install.sh)
  • Medium risk: network-variable-host (Memory/installers/install.sh:24)
  • Medium risk: network-variable-host (Memory/installers/install.sh:25)
  • Medium risk: runs-install-script-remote-exec (README.md:93)
  • Medium risk: runs-install-script-network-variable-host (scripts/install.sh)
  • Medium risk: network-variable-host (scripts/install.sh:116)
  • Medium risk: network-variable-host (scripts/install.sh:119)
  • Low risk: scripts-present (App/memmy-agent/src/skills/docx/scripts/render_docx.mjs)
  • Low risk: network-instructions (App/memmy-agent/src/skills/pdf/SKILL.md:32)
  • Low risk: network-instructions (App/memmy-agent/src/skills/update-setup/SKILL.md:87)
  • Low risk: scripts-present (Memory/installers/install.ps1)
  • Low risk: scripts-present (Memory/installers/install.sh)
  • Low risk: scripts-present (scripts/install.sh)
  • Low risk: scripts-present (scripts/internal/mac/build-dmg.sh)
  • Low risk: scripts-present (scripts/internal/shared/verify-package-version-lib.mjs)
  • Low risk: scripts-present (scripts/internal/shared/verify-package-version.mjs)
  • Low risk: scripts-present (scripts/test-literature-review-plugin-e2e.mjs)
  • Low risk: scripts-present (tests/smoke/local-agent-memory-smoke.ts)
  • Low risk: scripts-present (tests/smoke/memory-layer-smoke.ts)

Attribution and licence

First committed by
GitHub avatar of ZongYue99ZongYue99
Publisher
OpenMem (Community)
Source
MemTensor/memmy-agent · commit e640129
Licence
MIT License (A licence file holds personal contact details, so AutoGineer links to the files instead of copying them.)
Synced
Contributors
hijzy, endxxxx, syzsunshine219, Orsted290, Wang-Daoji

AutoGineer is not affiliated with or endorsed by ZongYue99 or OpenMem.

Install from source

A licence file holds personal contact details, so AutoGineer links to the files instead of copying them. MemTensor/memmy-agent at e640129