Agent skill

Memory Engine

by HelloRuru in HelloRuru/claude-memory-engine

Memory management system for Claude Code — Student Loop, Smart Context, Auto Learn, Session Handoff, Correction Cycle.

Install Memory Engine

skills CLI

$ npx skills add HelloRuru/claude-memory-engine --skill memory-engine -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI

$ gh skill install HelloRuru/claude-memory-engine memory-engine --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy

$ git clone --depth 1 https://github.com/HelloRuru/claude-memory-engine.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skill .claude/skills/memory-engine && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Paste into Claude Code

Install the "memory-engine" agent skill from https://github.com/HelloRuru/claude-memory-engine/tree/main/skill into .claude/skills/memory-engine/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "memory-engine", then confirm the skill loads.

Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

$skill-installer

$skill-installer install https://github.com/HelloRuru/claude-memory-engine/tree/main/skill

Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.

skills CLI

$ npx skills add HelloRuru/claude-memory-engine --skill memory-engine -a codex

Project install goes to .agents/skills/; add -g for ~/.codex/skills/.

GitHub CLI

$ gh skill install HelloRuru/claude-memory-engine memory-engine --agent codex

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/HelloRuru/claude-memory-engine.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skill .agents/skills/memory-engine && rm -rf skills-src

Use ~/.agents/skills/ instead of .agents/skills for a personal install.

Codex skills documentation · loads skills from .agents/skills/

Paste into Codex

Install the "memory-engine" agent skill from https://github.com/HelloRuru/claude-memory-engine/tree/main/skill into .agents/skills/memory-engine/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "memory-engine", then confirm the skill loads.

Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add HelloRuru/claude-memory-engine --skill memory-engine -a cursor

Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.

GitHub CLI

$ gh skill install HelloRuru/claude-memory-engine memory-engine --agent cursor

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/HelloRuru/claude-memory-engine.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skill .cursor/skills/memory-engine && rm -rf skills-src

Use ~/.cursor/skills/ instead of .cursor/skills for a personal install.

Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/

Paste into Cursor

Install the "memory-engine" agent skill from https://github.com/HelloRuru/claude-memory-engine/tree/main/skill into .cursor/skills/memory-engine/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "memory-engine", then confirm the skill loads.

Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

gemini skills install

$ gemini skills install https://github.com/HelloRuru/claude-memory-engine.git --path skill

--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.

skills CLI

$ npx skills add HelloRuru/claude-memory-engine --skill memory-engine -a gemini-cli

Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.

GitHub CLI

$ gh skill install HelloRuru/claude-memory-engine memory-engine --agent gemini-cli

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/HelloRuru/claude-memory-engine.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skill .gemini/skills/memory-engine && rm -rf skills-src

Use ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.

Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/

Paste into Gemini CLI

Install the "memory-engine" agent skill from https://github.com/HelloRuru/claude-memory-engine/tree/main/skill into .gemini/skills/memory-engine/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "memory-engine", then confirm the skill loads.

Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

GitHub CLI

$ gh skill install HelloRuru/claude-memory-engine memory-engine

Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).

skills CLI

$ npx skills add HelloRuru/claude-memory-engine --skill memory-engine -a github-copilot

Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.

Manual copy

$ git clone --depth 1 https://github.com/HelloRuru/claude-memory-engine.git skills-src && mkdir -p .github/skills && cp -r skills-src/skill .github/skills/memory-engine && rm -rf skills-src

Use ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.

GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/

Paste into GitHub Copilot

Install the "memory-engine" agent skill from https://github.com/HelloRuru/claude-memory-engine/tree/main/skill into .github/skills/memory-engine/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "memory-engine", then confirm the skill loads.

GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add HelloRuru/claude-memory-engine --skill memory-engine -a opencode

OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.

GitHub CLI

$ gh skill install HelloRuru/claude-memory-engine memory-engine --agent opencode

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/HelloRuru/claude-memory-engine.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skill .opencode/skills/memory-engine && rm -rf skills-src

Use ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.

OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/

Paste into OpenCode

Install the "memory-engine" agent skill from https://github.com/HelloRuru/claude-memory-engine/tree/main/skill into .opencode/skills/memory-engine/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "memory-engine", then confirm the skill loads.

OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

Facts

Skill name: memory-engine
GitHub stars: 134
Token cost: ~1.9k tokens
SKILL.md length: 804 words
Files: 3 (incl. references)
Skills in repo: 1
Repo updated: May 11, 2026
First seen: Oct 7, 2026
Licence: MIT

At a glance

Memory management system for Claude Code — Student Loop, Smart Context, Auto Learn, Session Handoff, Correction Cycle.

Works in 5 steps: Smart Context → Auto Learn (Pitfall Detection) → Student Loop (/reflect) → …
Tasks that involve Session handoff
SKILL.md covers Modules, Hooks, Commands (36 files, 18 pairs… and File Structure, plus 3 more sections
Calls git

What it does

Memory Engine is an agent skill from HelloRuru/claude-memory-engine. Memory management system for Claude Code — Student Loop, Smart Context, Auto Learn, Session Handoff, Correction Cycle. Triggered by memory commands (/save, /reflect, /handoff, /check) or memory-related questions. Not for general programming tasks.

Its SKILL.md is about 1.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 3 other files, including reference files (for example `references/auto-learn.md` and `references/smart-context.md`).

It sits in Agent Workflows, covering Session handoff and Agent memory. The repository describes itself as: Claude Code 的記憶系統 | A memory system built with hooks + markdown. Zero dependencies. The licence is MIT.

When your agent uses it

Tasks that involve Session handoff
Tasks that involve Agent memory

Example prompts

“/memory-engine”

Workflow steps

5 steps, taken from the step headings in SKILL.md.

Smart Context
Auto Learn (Pitfall Detection)
Student Loop (/reflect)
Session Handoff (/handoff)
Correction Cycle (/analyze)

What it can do on your machine

Read from SKILL.md and the folder at commit 04f085c. It shows what the files ask for, not the result of running them.

Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
- git
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Memory Engine loads about 1.9k tokens when it runs, and up to ~2.4k if it reads all its reference files. Until then it costs about 65 tokens; SKILL.md has 804 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it: ~65
When it runs · the whole SKILL.md, loaded when a task matches: ~1.9k
With references · SKILL.md plus every file in references/, read only if the agent opens them: ~2.4k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from HelloRuru/claude-memory-engine at commit 04f085c, republished under its MIT licence (© HelloRuru). 804 words, ~1,910 tokens.

Download SKILL.mdSave it as .claude/skills/memory-engine/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.

name: memory-engine
description: Memory management system for Claude Code — Student Loop, Smart Context, Auto Learn, Session Handoff, Correction Cycle. Triggered by memory commands (/save, /reflect, /handoff, /check) or memory-related questions. Not for general programming tasks.

Memory Engine v1.6

A memory and learning system for Claude Code, built with hooks and markdown.

Modules

1. Smart Context

Auto-detects the current project from your working directory and loads its memory.

Resolves per-project memory directory automatically — no hardcoded paths
Implementation: ~/.claude/scripts/hooks/session-start.js
Config: references/smart-context.md

2. Auto Learn (Pitfall Detection)

Detects pitfall patterns before context compression — retries, errors followed by fixes, user corrections.

Saves both the problem and the fix to ~/.claude/skills/learned/
Same mistake 3+ times across different days → suggests writing it into permanent rules
Implementation: shared-utils.js → detectPitfalls() + savePitfalls()

3. Student Loop (/reflect)

8-step learning cycle. First 3 steps are automatic (every session). Last 5 via /reflect.

Takes notes — records what was done, files changed, decisions made
Links them — tags the project, connects to previous notes
Spots patterns — scans for pitfall signals

There is no real "end" to a Claude Code conversation. Memory Engine saves at three points:

Every 20 messages (mid-session-checkpoint) — most reliable, self-counted
Before context compression (pre-compact) — pitfall detection runs here, context is fullest
When conversation ends (session-end) — best-effort, not guaranteed to fire

You don't need to remember to run any command before closing. 4. Review — read past 7 days, mark useful vs outdated 5. Refine — 4-question decision tree (Keep? Condense? Already covered? Delete as last resort) 6. Re-study — re-analyze cleaned data for buried patterns 7. Slim down — list removable items, wait for user confirmation 8. Wrap up — produce a report

4. Session Handoff (/handoff)

Pass context between Claude Code windows without losing progress.

/handoff saves a handoff-*.md file with progress, decisions, unfinished tasks
Next session auto-detects it on startup (session-start.js)
Mid-conversation detection via memory-sync.js
Each handoff is shown once, tracked by .handoff-read.json

5. Correction Cycle (/analyze)

Learns from user corrections — mistakes that don't show up in error logs.

Analyze (/analyze) — compare user's edits against rules, log missed rules, distill new ones
Correct (auto, before tasks) — scan the error list as a reminder
Reflect (/reflect step 6) — same mistake 3+ times → upgrade to hard rule

Hooks

Hook	File	What it does
SessionStart	session-start.js	Load last summary + project memory + pending handoffs + pitfall review + /reflect reminder
SessionEnd	session-end.js	Save summary + project index + backup (best-effort, may not fire)
PreCompact	pre-compact.js	Snapshot + pitfall detection + backup — the real safety net
UserPromptSubmit	memory-sync.js	Cross-session memory change detection + handoff detection
UserPromptSubmit	mid-session-checkpoint.js	Checkpoint every 20 messages
PreToolUse(Write)	write-guard.js	Sensitive file write warning
PreToolUse(Bash)	pre-push-check.js	Safety check before git push
—	memory-backup.sh	Bidirectional sync script (push/pull/sync) — v1.6

Shared logic between session-end.js and pre-compact.js is extracted into shared-utils.js.

Commands (36 files, 18 pairs EN + ZH)

EN	ZH	Function
/save	/存記憶	Save memory — auto-dedup, route to correct file
/reload	/讀取	Load memory into current conversation
/todo	/待辦	Cross-project task tracking
/backup	/備份	Push local memory to GitHub (`memory-backup.sh push`)
/sync	/同步	Bidirectional sync (`memory-backup.sh sync`) — v1.6 pull+push
/handoff	/交接	Session handoff — pass progress to next window
/diary	/回顧	Generate reflection diary
/reflect	/反思	Analyze pitfalls, find patterns
/learn	/學習	Manually save a pitfall
/analyze	/分析	Record corrections into error notebook
/correct	/訂正	Review error notebook anytime
/check	/健檢	Quick health scan
/full-check	/大健檢	Full audit
/memory-health	/記憶健檢	Memory file stats + capacity warnings
/memory-search	/搜尋記憶	Keyword search across all memory files
/recover	/想起來	Restore memory from GitHub (`memory-backup.sh pull`) — v1.6 auto-distribute
/compact-guide	/壓縮建議	When to compact and when not to
/overview	/全覽	List all available commands

Show full SKILL.md (256 more words)Show less

File Structure

text

~/.claude/
  scripts/hooks/
    session-start.js         # Load recall + smart-context + handoff
    session-end.js           # Save summary + backup (best-effort)
    pre-compact.js           # Pre-compression snapshot + pitfall detection + backup
    shared-utils.js          # Shared functions (transcript, pitfall, backup)
    memory-sync.js           # Cross-session sync + handoff detection
    mid-session-checkpoint.js
    write-guard.js
    pre-push-check.js
    memory-backup.sh         # Bidirectional sync (push/pull/sync) — v1.6
  sessions/
    {date}-{id}-session.md   # Session summaries
    {date}-{id}-compact.md   # Pre-compact snapshots
    diary/                   # Reflection diaries
  commands/
    save.md / 存記憶.md       # ...15 pairs
    handoff.md / 交接.md
  skills/learned/memory-engine/
    SKILL.md                 # This file
    references/
      smart-context.md       # Project detection rules
      auto-learn.md          # Pitfall detection rules

Core Principles

Test what you build

After creating any hook or feature, always run a test to confirm it works
"Written" does not mean "done" — see actual output before marking complete

/reflect 4-question decision tree

Does it serve the main purpose? → No → remove
Can it be condensed? → Yes → replace with refined version
Already covered by an existing rule? → Yes → don't duplicate
Delete is the last resort → only if Q1-Q3 don't apply

Troubleshooting

MEMORY.md over 200 lines with no warning

Cause: Didn't check line count before writing
Fix: Run wc -l before every write. Over 170 lines → move old content to standalone files first

Memory saved but next session doesn't know

Cause: Content written directly into MEMORY.md instead of creating a pointer
Fix: MEMORY.md holds pointers only. Details go in memory/*.md

Backup push fails

Cause: Conflicts in the memory repo or expired auth
Fix: Run git pull first, resolve conflicts if any

Memory from another device not showing up

Cause: Backup was push-only. Device A pushes to GitHub, but Device B never pulls it back into its local projects/ directories
Fix: v1.6 adds pull and sync modes to memory-backup.sh. The pull distributes global memory files from the backup repo into every local project directory that has a memory/ folder (newer file wins, won't overwrite local changes)
Root cause: Each device has different working directories, so projects/ paths differ. Global memory must be actively distributed, not just stored in one place

References

references/smart-context.md — Project detection rules
references/auto-learn.md — Pitfall detection rules + auto-learn flow

Files

SKILL.md and 2 other files (references) in skill of HelloRuru/claude-memory-engine.

SKILL.md
references/auto-learn.md
references/smart-context.md

Open the folder on GitHubat commit 04f085c

Compare with similar skills

Memory Engine next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Memory Engine compared with similar skills
Skill	Stars	Used in	Tokens	Auto-check	Licence	Repo updated
Memory Engine this skillHelloRuru/claude-memory-engine	134	—	~1.9k	Automated safety check: Pass	MIT	4 mo ago
Beads Task Memorygastownhall/beads	28k	—	~1.2k	Automated safety check: Pass	MIT	today
Session History Searchslopus/happy	24k	—	~3.1k	Automated safety check: Pass	MIT	today
Memori Long-Term MemoryMemoriLabs/Memori	17k	—	~2k	Automated safety check: Notes	Custom licence	4 days ago
Coding Agent Session Findercode-yeongyu/oh-my-openagent	70k	—	~2.8k	Automated safety check: Pass	Custom licence	today
Deja History Searchvshulcz/deja-vu	1.1k	1 repos	~1.3k	Automated safety check: Pass	MIT	yesterday

Similar skills

Beads Task Memory
gastownhall/beads
Tracks multi-session work with dependencies in the bd issue tracker so the agent can find ready tasks and recover its context after conversation compaction.
28k GitHub stars~1.2k tokensUpdated today
Agent WorkflowsAuto-check passed
Session History Search
slopus/happy
Searches past Claude Code, Codex and Cursor sessions and summarizes what was worked on, tried or decided, using extraction scripts instead of reading raw logs.
24k GitHub stars~3.1k tokensUpdated today
Agent WorkflowsAuto-check passed
Memori Long-Term Memory
MemoriLabs/Memori
Connects Claude Code to Memori Cloud for long-term memory, recalling stored context before substantive replies and saving new context afterward.
17k GitHub stars~2k tokensUpdated 4 days ago
Agent WorkflowsAuto-check: notes
Coding Agent Session Finder
code-yeongyu/oh-my-openagent
Finds, reads and reconstructs past coding-agent sessions across Codex, Claude, OpenCode, Senpi and many other local agent logs.
70k GitHub stars~2.8k tokensUpdated today
Agent WorkflowsAuto-check passed
Deja History Search
vshulcz/deja-vu
Searches this machine's own history of past AI coding sessions across every tool used on it, before work is redone or something is denied.
1.1k GitHub starsUsed in 1 repo~1.3k tokens
Agent WorkflowsAuto-check passed
Memori MCP Memory Usage
MemoriLabs/Memori
Teaches an MCP-connected agent when and how to call Memori's recall, summary, compaction, augmentation, feedback and quota tools to keep context across sessions.
17k GitHub stars~3.8k tokensUpdated 4 days ago
Agent WorkflowsAuto-check passed

Questions about Memory Engine

What does Memory Engine do?

Memory management system for Claude Code — Student Loop, Smart Context, Auto Learn, Session Handoff, Correction Cycle. Memory Engine is an agent skill from HelloRuru/claude-memory-engine. Memory management system for Claude Code — Student Loop, Smart Context, Auto Learn, Session Handoff, Correction Cycle.

When should I use Memory Engine?

Memory Engine fits situations like: tasks that involve Session handoff; tasks that involve Agent memory.

How do I install Memory Engine in Claude Code?

Run `npx skills add HelloRuru/claude-memory-engine --skill memory-engine -a claude-code`. Or copy the skill folder (skill in HelloRuru/claude-memory-engine) into .claude/skills/memory-engine in your project. Claude Code loads it when a task matches its description.

How do I install Memory Engine in Codex?

Run `npx skills add HelloRuru/claude-memory-engine --skill memory-engine -a codex`. Or copy the skill folder (skill in HelloRuru/claude-memory-engine) into .agents/skills/memory-engine in your project. Codex loads it when a task matches its description.

Can I use Memory Engine in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add HelloRuru/claude-memory-engine --skill memory-engine -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/memory-engine, .gemini/skills/memory-engine, .github/skills/memory-engine and .opencode/skills/memory-engine in your project.

What does Memory Engine need to run?

Going by SKILL.md and its folder, Memory Engine needs the command-line tools its instructions call (git).

Does Memory Engine access the network?

SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Memory Engine safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Memory Engine use?

Memory Engine is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Memory Engine use?

About 1.9k tokens (SKILL.md is roughly 7.6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 503 tokens, read only when the agent opens those files.

What are the alternatives to Memory Engine?

Skills that share tags, products or a category with Memory Engine: Beads Task Memory (gastownhall/beads, 28k stars), Session History Search (slopus/happy, 24k stars), Memori Long-Term Memory (MemoriLabs/Memori, 17k stars) and Coding Agent Session Finder (code-yeongyu/oh-my-openagent, 70k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Memory Engine?

HelloRuru (a GitHub user) maintains it in HelloRuru/claude-memory-engine, which has 134 GitHub stars. The repository was last updated on May 11, 2026.

Source: HelloRuru/claude-memory-engine on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.

Install Memory Engine

Facts

At a glance

What it does

When your agent uses it

Example prompts

Workflow steps

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Context cost

Safety

SKILL.md

Memory Engine v1.6

Modules

1. Smart Context

2. Auto Learn (Pitfall Detection)

3. Student Loop (/reflect)

4. Session Handoff (/handoff)

5. Correction Cycle (/analyze)

Hooks

Commands (36 files, 18 pairs EN + ZH)

File Structure

Core Principles

Test what you build

/reflect 4-question decision tree

Troubleshooting

MEMORY.md over 200 lines with no warning

Memory saved but next session doesn't know

Backup push fails

Memory from another device not showing up

References

Files

Compare with similar skills

Similar skills

Beads Task Memory

Session History Search

Memori Long-Term Memory

Coding Agent Session Finder

Deja History Search

Memori MCP Memory Usage

Related

Questions about Memory Engine