Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/HelloRuru/claude-memory-engineWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/helloruru/claude-memory-engine/learn)<a href="https://agentmods.dev/commands/helloruru/claude-memory-engine/learn"><img src="https://agentmods.dev/badge/commands/helloruru/claude-memory-engine/learn/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/helloruru/claude-memory-engine/learn"><img src="https://agentmods.dev/badge/commands/helloruru/claude-memory-engine/learn.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.00292 |
| Opus 5.5 | $0.00000 | $0.00117 |
| Sonnet 5.5 | $0.00000 | $0.00058 |
| Haiku 4.5 | $0.00000 | $0.00029 |
Grade A, and why
learn scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
/learn -- Auto-Learn from Mistakes
Automatically detect and save pitfall experiences during work sessions.
When to Trigger (Claude decides on its own)
- Pitfall resolved -- Tried the wrong approach, eventually found the fix
- Non-obvious workaround -- Browser quirks, library limitations, platform-specific behavior
- Path/environment traps -- Windows path issues,
~resolution errors, cross-platform differences - Deployment gotchas -- Hosting platform quirks, cache issues, DNS delays
- 3+ attempts to solve -- If it took 3+ tries, it's worth recording
What NOT to Learn
- Settings already documented in CLAUDE.md or project docs
- Common CSS/HTML patterns (easily searchable)
- One-off fixes that won't happen again
Save Format
Save to ~/.claude/skills/learned/{pattern-name}.md:
# {Descriptive Title}
**Date:** {today}
**Project:** {which project}
## Problem
{What happened, what context}
## Solution
{How it was fixed}
## Next Time
{What situation should remind you of this}
Behavior
- After saving, briefly tell the user: "Learned: {title}"
- Don't write a long explanation of why it was saved
- Learn quietly, don't interrupt the workflow
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 44 lines · 0 tokens per session scan A 3a4b23b99891
learn is a command published in the GitHub repository HelloRuru/claude-memory-engine (134 stars, last pushed 5mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 292 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
context-stats
Display context window usage and token statistics.
budget
Manage the token budget for Claude Code sessions. Model-aware — picks the right effective context window and prices per model. The session's REAL model is read from the transcript by the budget hook; config.model is only the fallback. Everything from Sonnet 4.6 up is 1M; Haiku 4.5 is 200K.
run
Execute prompt(s) from ./prompts/ with automatic archiving - for structured prompt-based work.
help
Explain Ralph Wiggum technique and available commands.
cancel-ralph
Cancel active Ralph Wiggum loop.
orchestrate
Do this task as an orchestrator. The aim is the same result at a fraction of the cost: in a long session, every token in this conversation is re-read on each later request, while a subagent's reading is discarded when it returns. The plugin's subagents run on Haiku 5.5, about 40× cheaper per token than Opus 5.5.