Explainer

Yapay zekâ agent belleği nedir? Kısa vadeli, uzun vadeli ve proje bağlamı

2026-08-27·8 min read·2026-08-27 tarihinde güncellendi

Yapay zekâ agent belleği, her istekte sıfırdan başlamak yerine sonraki görevlere taşınabilen bilgidir. Kısa vadeli bellek mevcut iş parçacığını, uzun vadeli bellek kalıcı gerçekleri, proje bağlamı ise tek bir çalışmanın dosya ve kararlarını tutar.

Üç bellek kapsamı

ScopeUse
Short-term memoryCurrent thread, tool results, and resumable work
Long-term memoryDurable preferences, rules, and lessons across sessions
Project contextSources, decisions, artifacts, and dependencies for one outcome

Kısa vadeli bellek: çalışma akışı

Short-term memory is the active state of a task. Keep recent decisions, unresolved questions, and tool results that affect the current deliverable. Summarize repetitive history and mark failed attempts so the agent does not repeat them blindly.

Uzun vadeli bellek: kalıcı gerçekler

Long-term memory should hold only stable, useful, expected, and properly scoped facts. Give each memory an owner, timestamp, correction path, and retention rule. Do not turn a one-off request into a permanent preference or store secrets by default.

Proje bağlamı: sınırları olan bellek

Project context keeps a brief, source files, approved facts, feedback, owners, and deadlines together. A project-specific claim should not silently become a global user preference. When a source changes, mark dependent work stale and show the provenance.

Getirme ve yönetişim

Storing information is separate from retrieving it. Check scope, relevance, freshness, authority, and conflicts before adding a memory to the current context. Current instructions and approved project sources should normally outrank old inferences.

Yaygın hatalar

  • Remembering everything increases noise, cost, and privacy exposure.
  • Wrong namespaces leak a project fact into unrelated work.
  • Stale memories override current instructions without dates or precedence.
  • Invisible memories cannot be corrected or deleted.
  • Memory is evidence, not a guarantee of truth.

Ottermind nasıl uyum sağlar?

Ottermind is an AI agent workspace that keeps goals, files, links, decisions, intermediate work, and follow-up tasks connected around their project. Use a bounded project, name the source of truth, review context before consequential output, and keep correction and forgetting available.

Sık sorulan sorular

Is chat history the same as agent memory?

No. Chat history is one source of short-term context. Agent memory also includes durable facts, summaries, checkpoints, and project records.

Does more memory make an agent smarter?

Not automatically. Better memory is scoped, fresh, traceable, editable, and relevant to the current task.

Sonuç

Short-term memory keeps a task moving, long-term memory carries carefully selected durable facts, and project context keeps one body of work coherent. Make the boundaries visible and design correction and forgetting from the start.

Research and disclosure: We reviewed the OpenAI Agents SDK memory guide, OpenAI Sessions guide, OpenAI context guide, and LangGraph memory documentation on August 27, 2026. These are implementation references, not an Ottermind benchmark.

Masaüstü ve mobil uygulamayı indir

Ottermind'e her zaman, her yerden eriş.

Bilgisayar