Retain, recall, and reflect on agent memories via self-hosted Hindsight.
npx skills add https://github.com/delorenj/00_ceiling --skill hindsight
Please help me install this Agent Skill. Skill: hindsight Source: https://github.com/delorenj/00_ceiling/tree/main/hindsight Command: npx skills add https://github.com/delorenj/00_ceiling --skill hindsight
This Skill provides persistent, structured memory for AI agents, enabling them to retain knowledge, recall context, and reflect on patterns over time, overcoming the inherent statelessness of most LLM interactions.
reflect to synthesize knowledge with configurable disposition and directives.Use the hindsight skill to retain the fact that the project deadline is April 15th.
High-intent search queries and answers about installing and using this skill.
Persistent memory for an LLM agent is achieved by retaining and recalling information across sessions. This enables the agent to remember user preferences, past task outcomes, and project details, overcoming LLM statelessness.
Contextual recall in AI agent memory retrieves relevant memories based on semantic similarity, keywords, graph connections, and temporal data. This allows the agent to access previously stored knowledge during current interactions.
Memory bank routing and automatic detection are configured through multi-bank routing architectures and automatic bank detection via git. This organizes and isolates agent memory retention by project context.
You can synthesize knowledge and enforce directives by using the reflect operation for agentic reasoning. Mental model management and directive enforcement allow the agent to process patterns and enforce rules.
The OpenClaw plugin supports automated memory capture and recall hooks. This integration facilitates seamless knowledge retention and contextual recall directly within the agent workflow.
Self-hosting agent memory and knowledge graphs requires managing your own infrastructure for persistence. You must handle the operational overhead of maintaining semantic similarity, graph connections, and temporal data retrieval.
freekmurzeExtract text and tables from PDFs using pypdf and pdfplumber.
lloydchangIndexes documents and performs semantic search across knowledge graphs and meeting transcripts.
aztr0nutzsStore persistent agent memory and build knowledge graphs with hybrid BM25, vector, and graph retrieval.
Clay-HHKStore embeddings and metadata for semantic and full-text retrieval.