Manage conversational memory with decay-based semantic retention and ChromaDB archival.
npx skills add https://github.com/wjlmrzd/OpenClaw-Wren --skill fluid-memory
Please help me install this Agent Skill. Skill: fluid-memory Source: https://github.com/wjlmrzd/OpenClaw-Wren/tree/main/skills/fluid-memory Command: npx skills add https://github.com/wjlmrzd/OpenClaw-Wren --skill fluid-memory
Fluid Memory solves the problem of unbounded, noisy conversational context by converting transient chat history into a semantic, decay-aware memory store that strengthens frequently accessed facts and archives or deletes low-value information automatically.
Ask the agent to remember that the user prefers cola so Fluid Memory will store that preference during the next OpenClaw memory flush.
High-intent search queries and answers about installing and using this skill.
Semantic memory with decay-based recall converts transient chat history into a vector store where frequently accessed facts are reinforced while stale or low-value information is automatically down-weighted, archived, or deleted over time.
You can store user preferences across sessions by asking the agent to remember the preference, which Fluid Memory commits to local ChromaDB vector storage during an OpenClaw memory flush event using incremental summarization.
Yes, you can automate archival and deletion of low-value conversational memories using a maintenance daemon that nightly consolidates archived items and hard-deletes long-archived entries based on configurable decay and boost thresholds.
OpenClaw supports incremental summarization for memory flush events by aggregating recent dialog into compact summaries and committing them to the semantic vector store, preventing unbounded and noisy conversational context.
The best way to manage unbounded conversational context in Python agents is applying a decay-aware semantic retention model that strengthens frequently accessed facts and archives obsolete details automatically using local vector storage.
FutureAtomsGenerate text embeddings via the Gemini API for semantic search and RAG.
Rwb3nAutomate triage and promotion of archived observations through a SCAN-TRIAGE-PROMOTE cycle.
kaifashraffStore, recall, and archive conversation memories with decay and semantic search.
qmuSearch archived tickets under .workaholic/tickets/archive and sort matches by relevance.
