ovos-memory-plugin-longterm: rolling summarization

LongTermMemory keeps a compact running summary of a long conversation. Every summarize_every exchanges it sends the oldest turns to an OpenAI-compatible chat endpoint, replaces them with the returned summary, and keeps only recent_window exchanges verbatim. The summary plus the recent window are persisted per session (JSON or SQLite).

Use it when a gist of the conversation is enough and you already have an LLM endpoint. It trades exact recall for a bounded, cheap-to-carry context. For exact recall of specific facts, prefer local-rag.

How it works

Session store (JSON or SQLite)
  rolling_summary  ← LLM summarize every N exchanges
  recent_window messages (verbatim)

build_conversation_context(utterance)
  [SYSTEM: system_prompt + rolling_summary]
  [... recent verbatim turns ...]
  [USER: utterance]

Configuration

Key	Type	Default	Description
api_url	str	http://localhost:8000/v1	OpenAI-compatible chat server
model	str	auto-detected	model for chat completions
summarize_every	int	6	exchanges accumulated before summarizing
max_summary_tokens	int	256	max_tokens for the summary request
recent_window	int	4	exchanges kept verbatim after each summary
backend	str	json	json or sqlite
db_path	str	~/.local/share/ovos/longterm_memory.{json,db}	persistence path
system_prompt	str	""	persona system prompt
request_timeout	int	30	HTTP timeout (s)

Persona wiring

{
  "name": "MyAssistant",
  "handlers": ["ovos-solver-openai-plugin"],
  "memory_module": "ovos-memory-plugin-longterm",
  "ovos-memory-plugin-longterm": {
    "api_url": "http://localhost:8000/v1",
    "model": "mistral",
    "summarize_every": 8,
    "recent_window": 4,
    "backend": "sqlite",
    "db_path": "~/.local/share/ovos/assistant_memory.db",
    "system_prompt": "You are a helpful assistant."
  }
}

---

Home · Local RAG →