By shuaiyy-ux
Context-triggered error prevention. Auto-captures bug patterns, auto-injects warnings when you're about to repeat them.
Context-triggered error prevention for Claude Code.
The problem isn't "knowing" a rule — it's "remembering it at the moment you're about to break it." Memory systems solve the former. memo-ultra solves the latter.
PreToolUse: Every time Claude is about to Edit/Write a file, memo-ultra scans the code content against a pattern library of past mistakes. If keywords match, a warning is injected into Claude's context at that exact moment.
Auto-capture: Every 3 conversation turns, memo-ultra scans the transcript for bug-fix signals and auto-extracts patterns into the library.
Manual learning: Use /memo-ultra:learn to teach it specific rules from past mistakes.
From GitHub:
git clone https://github.com/shuaiyy-ux/memo-ultra.git
claude --plugin-dir ./memo-ultra
Or permanent install in a Claude Code session:
/plugin marketplace add shuaiyy-ux/memo-ultra
/plugin install memo-ultra
| Command | Description |
|---|---|
/memo-ultra:learn | Teach a new error pattern |
/memo-ultra:patterns | View all patterns in the library |
/memo-ultra:forget | Remove a pattern |
Each pattern has:
["Modal", "animationType", "sheet"])manual (from /learn) or auto (from transcript analysis)A pattern triggers when 2+ keywords match the code being written.
memo-ultra/
├── .claude-plugin/plugin.json # Plugin manifest
├── hooks/hooks.json # Hook configuration
├── scripts/
│ ├── pre-check.sh # PreToolUse: pattern matching
│ ├── post-summarize.sh # Stop: auto-extract patterns
│ └── session-init.sh # SessionStart: load patterns
├── commands/
│ ├── learn/COMMAND.md # /memo-ultra:learn
│ ├── patterns/COMMAND.md # /memo-ultra:patterns
│ └── forget/COMMAND.md # /memo-ultra:forget
└── data/
├── patterns.example.jsonl # Example patterns (shipped with repo)
└── patterns.jsonl # Your pattern library (gitignored, auto-created)
Every other memory tool (claude-mem, etc.) solves "what did I do last session" — they bulk-inject summaries at session start. memo-ultra solves "I'm about to make the same mistake again" — it stays silent until the code you're writing matches a known error pattern, then surfaces the exact rule at that moment. No noise when irrelevant, instant recall when it matters.
Rules written in documents don't prevent mistakes. A rule only works if it appears at the exact moment someone is about to break it. memo-ultra puts past mistakes in front of Claude at the moment they're about to be repeated.
Modifies files
Hook triggers on file write and edit operations
No model invocation
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Executes directly as bash, bypassing the AI model
Executes directly as bash, bypassing the AI model
npx claudepluginhub shuaiyy-ux/memo-ultraDetects correction or long-term note-taking scenarios, runs lightweight scholar preflight on fresh tasks, keeps the agent in an explicit closed loop until the user confirms completion, escalates into Ascended Mode after repeated failed corrections or manual trigger, archives detailed `mistake` / `note` entries into project/global stores, and refreshes cache-like distilled memory.
I (Claude) act as you - Build knowledge base from your notes.
Self-learning system for Claude Code that captures corrections and updates CLAUDE.md automatically
Curated persistent memory for Claude Code. Write gate prevents bloat — only behavior-changing facts get saved. Tiered architecture: daily logs, structured registers, and auto-loaded working memory.
Self-improving Claude Code plugin — learns from corrections across sessions via reflexio
Multi-model consensus engine integrating OpenAI Codex CLI, Gemini CLI, and Claude CLI for collaborative code review and problem-solving.