Store and retrieve agent memories via pgvector in Supabase.
npx skills add https://github.com/teodorboev/socialai --skill shared-memory-teodorboev
Please help me install this Agent Skill. Skill: shared-memory Source: https://github.com/teodorboev/socialai/tree/main/.opencode/skills/shared-memory Command: npx skills add https://github.com/teodorboev/socialai --skill shared-memory-teodorboev
This Skill provides a persistent, searchable memory layer for AI agents, allowing them to learn from past interactions, decisions, and outcomes, thereby enabling more intelligent and context-aware behavior over time.
Use the shared-memory skill to store a memory about a successful marketing campaign.
High-intent search queries and answers about installing and using this skill.
To add persistent memory to an AI agent using Supabase, you implement a vector database layer with pgvector that stores and retrieves historical context, enabling stateful agent behavior over time.
Semantic recall for AI agents is a memory mechanism that stores past interactions and retrieves them based on semantic similarity, ensuring relevant historical context is provided before execution to inform decisions.
Yes, you can use pgvector to store human feedback and past performance by saving execution outcomes into the database, allowing the agent to learn from past topics and optimize future tasks.
The best way to make an AI content agent avoid repeating topics is to store historical engagement metrics in a vector database, recalling previously successful posts to inform new content generation and prevent repetition.
Yes, an agent memory layer integrates with a base agent architecture by performing semantic recall before execution to gather context, and storing new interaction data after execution completes.
CleanExpoGenerate pgvector embeddings and run similarity or hybrid search in Supabase/PostgreSQL.
Mdx2025Store and retrieve OpenClaw agent memories via PostgreSQL pgvector semantic search.
wrsmith108Automate Claude Code hooks with MCP integration for multi-agent workflows.
semaes111Orchestrate seven memory layers for AI agents across sessions.
