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.
npx agentmods add commands/runtimenoteslabs/memory-layer/remembergit clone --depth 1 https://github.com/runtimenoteslabs/memory-layerWhat 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 | $0.00009 | $0.00470 |
| Opus 5 | $0.00005 | $0.00235 |
| Sonnet 5 | $0.00002 | $0.00094 |
| Haiku 4.5 | $0.00001 | $0.00047 |
Grade A, and why
remember 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 2d 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.
How it starts
The opening of the file, as written. The whole thing — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Remember Command
Store a piece of knowledge in the Memory Layer for future retrieval. Memories are persisted across sessions and become smarter over time through outcome-based learning.
Usage
/remember <content> [-c <category>]
Categories
| Category | Use For |
|---|---|
architecture |
System structure, component relationships, design patterns |
convention |
Coding standards, naming patterns, style rules |
decision |
Technical choices and their rationale |
pattern |
Reusable solutions, idioms, best practices |
gotcha |
Pitfalls, edge cases, things that don't work as expected |
workaround |
Temporary fixes, hacks, technical debt |
troubleshooting |
Error solutions, debugging steps, fixes |
command |
Useful CLI commands, scripts, one-liners |
preference |
User preferences for tools, formatting, workflow |
Examples
# Store a coding convention
/remember "Always use snake_case for Python variables" -c convention
# Store a gotcha about the project
/remember "The auth service rate limits at 100 req/min" -c gotcha
# Store a technical decision with rationale
/remember "We chose PostgreSQL over MySQL for JSON support and better concurrency" -c decision
# Store a pattern
/remember "Use the repository pattern for all database access" -c pattern
# Store a troubleshooting tip
/remember "Error 'connection refused' usually means Redis isn't running" -c troubleshooting
# Store without category (auto-categorized)
/remember "npm run build:prod for production builds"
Implementation
Parse the arguments and store via CLI:
mem add "$ARGUMENTS"
After storing, confirm to the user what was saved and its assigned ID. The memory starts with an outcome score of 0.0 and will be adjusted based on feedback.
Tips
- Be specific and actionable in your memories
- Include context about when/why the advice applies
- Use consistent terminology to improve retrieval
- Record outcomes with
/outcometo improve future suggestions
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.
- 2d ago First seen · 68 lines · 9 tokens per session scan A bf29eb7cbc25
remember is a command published in the GitHub repository runtimenoteslabs/memory-layer (10 stars, last pushed 3mo ago), licensed MIT. It adds 9 tokens to every session and 470 once invoked, about $0.0000 per session on Opus 5. 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 2026-08-31.
Other commands, from other repositories
memory-status
Check the Pseudolife-MCP memory daemon and report bank health.
dream
Copy to .claude/commands/dream.md in any project to get /dream. -->.
tokenless-stats
Show Tokenless compression statistics.
evolve-lite-subscribe
Add a shared guidelines repo (read-scope subscription or write-scope publish target) to the unified repos list.
cb-search
코드브레인 BM25 검색 — stale 자동 refresh.
graphify
Turn your vault into a clustered knowledge graph with HTML and JSON outputs.