Back to Home
smart_toy

AI Agent Memory & Knowledge

About AI Agent Memory & Knowledge

These skills equip AI agents with core memory and knowledge management, enabling persistent learning, efficient context understanding, and optimized decision-making. It covers various storage solutions from short-term conversational memory to cross-session long-term memory, supporting knowledge ingestion, organization, synchronization, and retrieval, such as managing notes via Obsidian or utilizing vector databases. Ideal for developers and AI engineers building intelligent assistants, code Copilots, or any AI agent requiring memory of user preferences, complex context comprehension, and continuous evolution.

Subcategory navigation

Categories:

All AI Agent Memory & Knowledge Skills

50 / 151 skills
para

para-memory-files

by @getpaperclipai

Implements a three-layer file-based memory system (knowledge graph, daily notes, tacit knowledge) using the PARA method, enabling persistent cross-session context and efficient semantic retrieval for AI agents. Ideal for agents that need long-term recall of user preferences, project details, and personal knowledge bases.

4.9(120)
299.6K
technical-writing

ubiquitous-language

by @mattpocock

Automatically extract domain terms from conversations to form a unified glossary, eliminating homonymic ambiguities, and save as a local file for team reference.

4.8(33)
204.5K
obsidian-markdown

obsidian-cli

by @kepano

Interact with Obsidian vaults via Obsidian CLI to read, create, search, and manage notes, tasks, and properties.

4.7(285)
75.5K
ai-agent

dbs-save

by @dontbesilent2025

dbs-save is a state-saving tool for dbskill, designed to structure and archive key conclusions, user-rejected directions, and next steps from AI diagnostic sessions into local Markdown files. It elevates diagnostics from one-off consultations to a cumulative 'personal doctor' model, ensuring user history and decisions are not lost. Through project isolation and automatic title extraction, it efficiently manages diagnostic states, providing a foundation for future restoration and reporting, thereby enhancing the continuity and user experience of AI diagnostics.

4.6(120)
20.8K
state-management

dbs-restore

by @dontbesilent2025

dbs-restore is a state restoration tool designed to help users resume previous diagnostic sessions from saved archives. It locates and loads specific projects or the latest archives based on user commands, presenting the conclusions, hypotheses, and next steps from the last diagnosis. By restoring context, users can seamlessly continue unfinished work, avoid re-explaining, and improve diagnostic efficiency.

4.6(120)
20.8K
decision-making

dbs-decision

by @dontbesilent2025

dbs-decision is a personal decision system AI that doesn't make decisions for the user. Instead, it helps build a local knowledge engineering system by structuring facts, patterns, states, and open questions into corresponding Markdown files. This ensures critical decision-making information is persistent, traceable, and supports future discussions and reviews, thereby enhancing decision quality and efficiency.

4.6(120)
18.5K
project-management

planning-with-files-zh

by @othmanadi

File planning system, uses persistent Markdown files as disk-based working memory, supporting multi-session context recovery and progress tracking.

4.6(20)
17.9K
content-structuring

dbs-content-system

by @dontbesilent2025

This AI skill builds a sustainable, reusable, traceable, and re-combinable content structuring system for large local content assets. It transforms scattered articles, tweets, and case studies into minimal semantic content units, establishing relationships, deduplication indexes, and topic maps. This provides AI agents with an efficient knowledge storage and retrieval foundation, significantly enhancing the utilization of content assets.

4.6(120)
17.3K
langchain

deep-agents-memory

by @langchain-ai

Manages AI agent memory, offering short-term (in-thread), long-term (cross-session), and hybrid storage solutions to ensure information persistence and routing.

4.6(46)
16.8K
langgraph

langgraph-persistence

by @langchain-ai

Saves and loads graph states via Checkpointer, manages sessions using thread IDs, and provides cross-thread storage for user preferences.

4.5(24)
15.6K
data-caching

memory-merger

by @github

Merges mature experiences from domain memory files into instruction files, supporting global, user, and workspace scopes.

4.5(348)
12.9K
session-history

session-history

by @rohitg00

This skill provides a clear timeline of recent project sessions, enabling users to quickly review past work. It's particularly useful when a user asks 'what did we do last time', 'session history', or wants an overview of previous work. By presenting concise session records, this Skill significantly enhances work efficiency and improves contextual understanding for ongoing projects.

4.5(120)
12.2K
memory-management

forget

by @rohitg00

This skill enables users to delete specific observations from an AI agent's memory, ensuring data privacy and memory accuracy. By providing an explicit confirmation mechanism before deletion, it prevents accidental data loss and helps the agent maintain a clean, relevant knowledge base, thereby improving its decision-making quality and user trust.

4.5(120)
12.1K
agent-memory

recap

by @rohitg00

This skill summarizes the last N agent sessions for the current project, grouped by date, with highlight observations per session. It helps users quickly review project progress and understand the agent's working context, thereby improving decision-making efficiency and project management capabilities. Particularly useful for scenarios requiring a quick overview of recent agent activities or project status.

4.5(120)
12.1K
agent-memory

commit-context

by @rohitg00

This skill traces a code file, function, or specific line back to the AI Agent session that produced its current commit. When users need to understand 'why this code is here,' 'what the Agent was doing during the change,' or require context for a specific location in the codebase, this Skill provides crucial historical background and decision-making basis, significantly improving code comprehension and troubleshooting efficiency.

4.5(120)
12.0K
contextual-ai

context7

by @intellectronica

Retrieve and provide the latest documentation for software libraries, frameworks, and APIs, ensuring developers get accurate, real-time technical information.

4.5(136)
11.0K
contextual-ai

opencontext

by @supercent-io

This skill provides AI assistants with persistent memory and context management capabilities, eliminating the need for repetitive explanations and enabling the construction of smarter, more efficient AI systems.

4.5(420)
10.5K
knowledge-management

dbs-knowledge

by @dontbesilent2025

This skill transforms local folders into an AI Agent-ready knowledge base, automatically creating navigation files (SOURCE_OF_TRUTH.md), managing version conflicts, and supporting search, storage, and health checks. It helps developers, teams, or individuals organize documents and data structurally, enabling AI agents to quickly locate and use correct information, eliminating manual searching and version chaos.

4.5(120)
10.2K
similarity-search

similarity-search-patterns

by @wshobson

Familiarize with similarity search patterns, utilizing intelligent automation and multi-agent orchestration techniques to efficiently process and analyze large-scale data, improving retrieval efficiency.

4.5(74)
9.7K
ai-memory

remember

by @github

Transform lessons learned into memory instructions organized by domain (global or workspace), recording and managing knowledge through specific syntax.

4.5(295)
9.3K
wechat

baoyu-wechat-summary

by @jimliu

This skill intelligently summarizes WeChat group chat highlights into structured, readable digests, offering an optional "roast" version for a humorous take. It meticulously maintains per-group chat history and individual user profiles across multiple runs, allowing users to effortlessly review past discussions, understand participant dynamics, and efficiently extract key information from their busy WeChat groups.

4.5(120)
9.3K
ai-agent-memory

agentmemory-mcp-tools

by @rohitg00

This skill provides a detailed map of all agentmemory MCP tools, outlining their functions and parameters. It assists users in selecting the appropriate memory tool, clarifying tool names or arguments, and understanding what agentmemory can achieve via MCP. This Skill significantly streamlines AI agent memory management and knowledge retrieval, ensuring correct and efficient tool utilization.

4.5(120)
9.3K
rest-api

agentmemory-rest-api

by @rohitg00

This skill provides the HTTP REST API surface for agentmemory, serving as the primary protocol for interacting with the memory server. It enables AI agents to store and retrieve memories via standard HTTP, ensuring efficient knowledge base management and access in various scenarios, such as when a specific protocol is unavailable or when integrating with incompatible hosts. This significantly enhances the agent's adaptability and data interaction capabilities.

4.5(120)
9.3K
conversational-ai

remembering-conversations

by @obra

After exploring code or encountering issues, help users think through the best approaches or solutions, providing conversational memory and guidance.

4.5(276)
9.2K
agent-memory

agentmemory-architecture

by @rohitg00

This skill details the architecture of agentmemory, including its iii engine primitives, storage model, ports, and viewer. It explains how memories are stored, retrieved, and managed, covering hybrid retrieval mechanisms and the memory lifecycle. Developers can leverage this skill to deeply understand agentmemory's internal workings, enabling better system extension, troubleshooting, or optimization of memory strategies, ensuring AI agents efficiently manage and utilize knowledge.

4.5(120)
9.2K
agent-memory

agentmemory-hooks

by @rohitg00

This skill automatically captures observations throughout an AI agent's session lifecycle, enabling automated memory management. It helps developers understand how memory is captured, debug missing observations, and fine-tune what gets recorded. This enhances the agent's autonomous learning and decision-making capabilities by ensuring comprehensive and accurate memory retention.

4.5(120)
9.2K
interactive-programming

remember-interactive-programming

by @github

A micro-prompt reminding the agent of its role as an interactive programmer, excelling in REPL environments like Clojure.

4.5(292)
9.0K
memory-search

mem-search

by @thedotmack

Searches historical work records across all sessions, providing a simple workflow for searching, filtering, and retrieving past information, improving work efficiency.

4.5(70)
8.3K
agent-skill

khazix-skills

by @KKKKhazix

After each task is completed, project documents, CLAUDE.md, and Agent memory are automatically aligned, and a change summary is generated. This ensures information consistency and prevents the Agent's efficiency from declining due to outdated information.

4.5(322)
8.0K
project-management

planning-with-files-zht

by @othmanadi

Use persistent Markdown files as disk-based working memory, supporting cross-session context recovery.

4.5(20)
7.4K
documentation

hindsight-docs

by @vectorize-io

Provides comprehensive Hindsight documentation for AI agents, aiding in understanding and analyzing AI behavior, optimizing decision-making, and enhancing system transparency.

4.5(20)
7.3K
rag-engine

agent-platform-rag-engine-management

by @google

This skill offers comprehensive guidance on interacting with the Agent Platform RAG Engine. It enables developers to efficiently manage RAG engines using the Python SDK, covering operations such as listing and inspecting corpora, retrieving contexts, and generating answers based on retrieved information. This capability is crucial for building and maintaining powerful knowledge storage and retrieval systems for AI agents, thereby improving their decision-making and response accuracy.

4.5(120)
6.8K
ai-memory

timeline-report

by @thedotmack

A persistent memory compression system for Claude Code. It efficiently manages and compresses contextual information, helping AI agents overcome long text limitations for deeper understanding and coherent conversations. This enhances AI's performance and memory in complex code and extended interactions.

4.4(20)
6.3K
planning

planning-with-files-es

by @othmanadi

This skill provides AI agents with a file-based persistent planning system to organize, track, and execute complex multi-step tasks. By creating task plans, findings, and progress files, it helps AI maintain context, avoid repetitive errors, and efficiently complete projects.

4.4(120)
6.3K
task-planning

planning-with-files-ar

by @othmanadi

This is a Manus-style file planning system that provides persistent memory for AI agents by creating `task_plan.md`, `findings.md`, and `progress.md` files. It effectively organizes and tracks the progress of complex tasks, supports automatic session recovery, and significantly enhances the efficiency and reliability of task planning, project management, and progress tracking.

4.4(120)
6.2K
security-auditing

audit-context-building

by @trailofbits

Manages Claude's thinking process during the audit context building phase, enabling ultra-fine pure context mode.

4.4(20)
5.8K
book

book-to-skill

by @virgiliojr94

Converts a technical book (PDF or EPUB) into a structured Claude Code skill — extracting frameworks, mental models, principles, techniques, and anti-patterns the author crystallized. Use when the user wants to study a book through Claude, apply an author's frameworks while working, or build a reusable knowledge base from any PDF or EPUB.

4.4(120)
5.3K
knowledge-management

knowledge-agent

by @thedotmack

This skill builds and queries AI-powered knowledge bases from historical observation data, providing customized "brains" for AI agents. It allows users to filter information by project, concept, or date range, quickly creating focused knowledge collections on specific topics. This helps agents remember user preferences, understand complex situations, and efficiently answer questions about past work patterns or specialized domains.

4.4(120)
5.2K
search-algorithms

search-strategy

by @anthropics

The core intelligence for enterprise search, transforming single queries into effective search strategies. An AI Agent Skill to improve work efficiency and automation capabilities.

4.4(20)
5.1K
history

joseon-sillok-search

by @nomadamas

Searches historical documents of the Joseon Dynasty Annals, supporting keyword search, dynasty filtering, year filtering, and providing Korean translation and original text comparison.

4.4(8)
4.9K
ai-agent-memory

how-it-works

by @thedotmack

The claude-mem skill explains how it captures AI agent activities through compressed observations and automatically injects relevant memories into subsequent sessions to provide context. This allows the AI to remember project details and decisions, avoiding repetitive codebase explanations or rediscovering information, significantly enhancing development efficiency and coherence.

4.4(120)
4.6K
github-copilot

copilot-spaces

by @github

Uses Copilot Spaces to provide project-specific context for conversations, loading relevant information from shared spaces.

4.4(49)
4.5K
context-management

ck

by @affaan-m

Manage Claude Code's context state, supporting saving, restoring, and switching different conversation contexts, avoiding information loss in long conversations.

4.4(24)
4.5K
knowledge-management

knowledge-ops

by @affaan-m

Manage multi-layered knowledge systems, performing knowledge ingestion, organization, synchronization, and retrieval across multiple stores, supporting workspaces like code repositories, GitHub, and Linear.

4.4(27)
4.3K
knowledge-base

kb-retriever

by @conardli

This skill acts as an intelligent retrieval and Q&A assistant for local knowledge base directories. It navigates through hierarchical indexes and, for file types like PDF/Excel, mandates learning specific processing methods before performing progressive retrieval, avoiding full file loading. By combining tools like grep, Read, pdfplumber, and pandas, it efficiently extracts information from the knowledge base to answer user queries related to data lookup and information retrieval.

4.4(120)
4.2K
knowledge-management

llm-wiki

by @ar9av

Build a persistent knowledge base, distilling information once and keeping it updated, rather than re-deriving it with each query.

4.4(10)
3.9K
ontology-engineering

ontology

by @sundial-org

In data and AI, this skill provides typed knowledge graphs for building structured agent memories and composable knowledge representations, enhancing AI system understanding and reasoning.

4.4(60)
3.7K
knowledge-retrieval

wiki-query

by @ar9av

Perform knowledge retrieval against a compiled Obsidian wiki, returning pre-synthesized cross-referenced knowledge.

4.4(21)
3.4K
knowledge-management

neat-freak

by @KKKKhazix

This skill rigorously cleans up and synchronizes project documentation (e.g., CLAUDE.md, README.md, docs/) and AI agent memory after each session. It ensures the project's knowledge base remains clean, accurate, and newcomer-friendly, preventing stale documents and conflicting memories, thereby facilitating team collaboration and smooth handoffs.

4.4(120)
3.3K
knowledge-management

wiki-status

by @ar9av

This skill audits and tracks the current status of an Obsidian knowledge base. By comparing filesystem sources with an ingestion manifest, it identifies new, modified, or deleted content. This helps users decide between incremental updates or full rebuilds, ensuring the AI agent's knowledge memory is always current, enhancing data management efficiency and knowledge base accuracy.

4.4(120)
3.3K
🎁 Agent Knowledge Cards
Survey