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 skills add fthsrbst/mnema --skill hub-memorygit clone --depth 1 https://github.com/fthsrbst/mnemaWrote 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/skills/fthsrbst/mnema/hub-memory)<a href="https://agentmods.dev/skills/fthsrbst/mnema/hub-memory"><img src="https://agentmods.dev/badge/skills/fthsrbst/mnema/hub-memory/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/skills/fthsrbst/mnema/hub-memory"><img src="https://agentmods.dev/badge/skills/fthsrbst/mnema/hub-memory.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 | $0.00066 | $0.01495 |
| Opus 5.5 | $0.00026 | $0.00598 |
| Sonnet 5.5 | $0.00013 | $0.00299 |
| Haiku 4.5 | $0.00007 | $0.00150 |
Grade A, and why
hub-memory 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.
How it starts
The opening of the file, as written. The whole thing — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hub Hafıza Kullanımı — Katmanlı Bağlam
Hub, tüm cihazlardaki tüm agentların paylaştığı hafızadır. Sen kapanınca bağlamın silinir; hub silinmez. Bağlam üç katmandır: otomatik gelen (bridge/recall), görev başında çektiğin (project_get/memory_search), çalışırken yazdığın (memory_save/session_log). Doğru davranış katmanına göre değişir.
Katman 1 — Otomatik gelen (okuman yeter)
<hub-bridge>(oturum başı): aktif projenin map'i + son oturum özeti. Kaldığın yeri buradan al. Map'in gerçekle çeliştiğini görürsen görevin ortasında bileproject_updateile düzelt.<hub-recall>(mesaj başı): mesajla yüksek benzerlikli az sayıda kayıt. Sistem bilinçli olarak az ve isabetli enjekte eder; boş olması "hafızada yok" demek değildir — şüphen varsa Katman 2'ye in.
Katman 2 — Görev başında sen çek
- Projede çalışıyorsan ve bridge gelmediyse
project_get(name): özet, stack, kararlar, odak, sıradaki adımlar + kod haritası (architecture/modules/entry_points/commands). - "Neden X kullanıyoruz", "bunu daha önce nasıl çözmüştük" →
memory_search(query). - Doküman/öğrenme notu/araştırma arşivi →
rag_search(query). - "Nerede kalmıştım" →
session_recent(project).
Katman 3 — Çalışırken yaz
| Durum | Aksiyon |
|---|---|
| Teknik karar alındı (kütüphane, mimari, yaklaşım) | memory_save type=decision — gerekçeyi mutlaka yaz |
| Zor bir bug çözüldü | memory_save type=howto — belirti, kök neden, çözüm |
| Kullanıcı tercih belirtti (stil, araç, workflow) | memory_save type=preference |
| Tekrar lazım olacak komut/kurulum/config | memory_save type=howto |
| Uzun araştırma/öğrenme çıktısı, doküman, talimat/prompt metni | rag_add — memory DEĞİL (öğrenme notu: project='learning', uri='learning/') |
Kayıt kalite kuralları
- Ölçüt: "Başka bir cihazdaki agent 2 hafta sonra bundan faydalanır mı?" Hayırsa kaydetme (oturuma özel detay, koddan/git'ten okunabilen şey, geçici durum).
- Boyut: memory "bir bakışta okunur tek bilgi"dir. Gövde ~1500 karakteri aşıyorsa büyük ihtimalle doküman yazıyorsun →
rag_add. - project alanı:
project_list'teki kanonik ad. Makine/cihaz adı (fatih-pc gibi) proje DEĞİLDİR — cihaza özgü bilgiye tags ekle. Yeni gerçek proje açılıyorsa önceproject_updateile map oluştur. - importance: varsayılan 1 çoğu kayıt için doğrudur. 2 = aylar sonra bile öne geçmesi gereken kritik karar/tercih (nadir!). 0.5 = önemsiz detay. Enflasyon skorlamayı bozar.
- Dedup:
memory_savebenzer kayıt uyarısı dönerse ciddiye al — mevcut kaydımemory_updateile zenginleştir, yeni açtığınımemory_deleteile geri al. - Yanlışlanan bilgiyi gördüğün an düzelt (
memory_update) veya sil (memory_delete). Çelişkili hafıza, hafızasızlıktan kötüdür.
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 · 45 lines · 66 tokens per session scan A 749a826a2160
hub-memory is a skill published in the GitHub repository fthsrbst/mnema (2 stars, last pushed 1mo ago), licensed MIT. It adds 66 tokens to every session and 1,495 once invoked, about $0.0003 per session on Opus 5.5. 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 skills, from other repositories
memorywhale
Query and write durable debugging memory recorded by MemoryWhale. Use when debugging a failure that may have happened before, when you need the exact error/flags/output from an earlier attempt, when the user asks "how did we fix this last time?", or once you've figured out why something failed / how a fix worked and…
recall
Use Recall to search, inspect, continue, export, resume, or share indexed AI coding sessions. Trigger for project-history lookup, recent work from other agents, file history, unfinished-session continuation, and published session-page management.
reflect
Use Recall Reflect to review AI coding workflow history as a conversation-first timeline and discuss observed patterns before changing behavior. Trigger when the user asks to reflect on a project, review AI coding workflow, inspect workflow friction, discuss repeated corrections, review handoffs, or explore possible…
memorywhale-evidence
Use MemoryWhale debugging evidence when the user requests recall or a recurring failure may have relevant recorded history.
memorywhale-debugging
MemoryWhale debugging; compiler failures; terminal diagnostics.
daily-brief-loop
Use only when the user explicitly asks to run, sync, advance, or persist Mora's durable daily brief. Executes the once-per-day, human-checkpointed NO-SEND operator loop, performs configured source syncs, writes the dated artifact and run log, and advances the brief watermark.