AI Agent Memory Paradigms

AI Agent Memory Paradigms is a skill for Claude Code, Codex from j4flmao/agent-skills. It costs 25 tokens per session (647 once invoked), scanned A, original, MIT.

A guide to three kinds of memory for AI agents: working memory for the current task, semantic memory for stored facts, and episodic memory for past events. Together, these describe how an agent can retain information over time.

In plain words
What is it for?
Use it to plan memory systems for agents, including current context, knowledge bases, and records of previous interactions or experiences.
Why use it?
An agent that only sees the current conversation can lose important context during long tasks. Separating memory types helps explain what information should remain temporary or be stored.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it to plan memory systems for agents, including current context, knowledge…

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/j4flmao/agent-skills/memory-systems
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add j4flmao/agent-skills --skill memory-systems
Clone the repo
git clone --depth 1 https://github.com/j4flmao/agent-skills

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for AI Agent Memory Paradigms

README.md
[![agentmods](https://agentmods.dev/badge/skills/j4flmao/agent-skills/memory-systems.svg)](https://agentmods.dev/skills/j4flmao/agent-skills/memory-systems)
Your own site
<a href="https://agentmods.dev/skills/j4flmao/agent-skills/memory-systems"><img src="https://agentmods.dev/badge/skills/j4flmao/agent-skills/memory-systems.svg" alt="Measured on agentmods" height="20"></a>
Per session 25 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 647 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00025 $0.00647
Opus 5 $0.00013 $0.00324
Sonnet 5 $0.00005 $0.00129
Haiku 4.5 $0.00003 $0.00065

Measured 6d ago against content hash 708a2673e807, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

AI Agent Memory Paradigms scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/ai/ai-agents/memory-systems/SKILL.md · 48 lines

How it starts

The opening of the file, as written. The whole thing — 48 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Memory Paradigms: The Architecture of Continuity

An AI Agent without memory is temporally blind; its existence is constrained to the immediate context window. True autonomy requires a multi-layered memory architecture to simulate the continuity of consciousness and enable long-horizon coherence. This architecture is strictly categorized into three fundamental tiers.

1. Working Memory (The Context Window)

The immediate, transient cognitive space. This is the absolute limit of the agent's active reasoning capacity, defined by the underlying LLM's context window.

  • Nature: Highly volatile, exact retrieval, strictly bounded.
  • Function: Holds the current goal, immediate environmental state, recent observations, and the active reasoning trace.
  • First Principle: Context is a scarce resource. Information must be aggressively compacted or evicted to prevent attention degradation and catastrophic forgetting of immediate instructions.

2. Semantic Memory (The Knowledge Base)

The vast, static repository of facts, concepts, and externalized knowledge. This is typically implemented via dense vector embeddings and approximate nearest neighbor search.

  • Nature: Persistent, associative retrieval, theoretically unbounded.
  • Function: Provides domain-specific context injected dynamically into Working Memory based on semantic proximity to the current cognitive state.
  • First Principle: Semantic memory lacks temporal coherence. It provides "what is", not "what happened". It is highly dependent on embedding quality and chunking strategy to minimize retrieval noise.

3. Episodic Memory (The Experiential Ledger)

The chronological sequence of past events, actions, and outcomes. This is the agent's autobiographical memory, essential for complex reasoning across temporal gaps and learning from past failures.

  • Nature: Persistent, temporal/sequential retrieval.
  • Function: Enables reflection, trajectory evaluation, and the synthesis of abstract rules from concrete experiences.
  • First Principle: Raw logs are not episodic memory. True episodic memory requires the distillation of continuous state transitions into discrete, semantic narratives ("experiences") that can be queried by similarity or sequence.

Read the full file on GitHub · 48 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago First seen · 48 lines · 25 tokens per session scan A 708a2673e807

Subscribe to this mod's changes

AI Agent Memory Paradigms is a skill published in the GitHub repository j4flmao/agent-skills (21 stars, last pushed yesterday), licensed MIT. It adds 25 tokens to every session and 647 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

media-ingest

Ingest video, audio, PDF, book, screenshot, and GitHub repo content into the brain. Multi-format handling with entity extraction and backlink propagation. Covers video-ingest, youtube-ingest, and book-ingest subtypes.

garrytan/gbrain · 52 tokens

mem0-oss-to-platform

Plan and then execute a migration of a project from the mem0 open-source / self-hosted SDK (the local Memory class) to the mem0 Platform / hosted / managed SDK (the MemoryClient class). Use this whenever a developer wants to move, switch, or migrate their mem0 usage off OSS/self-hosted to the hosted API — e.g.…

mem0ai/mem0 · 273 tokens

Cortex

Operate Cortex, the LifeOS memory system — the typed Knowledge Archive (People, Companies, Ideas, Research with typed related: links) plus recall of prior work sessions, ISAs, and conversations. Search, add, harvest, develop, ingest, distill, graph-navigate, recall. USE WHEN cortex, knowledge, knowledge base, search…

danielmiessler/LifeOS · 196 tokens

memory

Use when the user asks to remember, recall, forget, update, search, or inspect durable OpenSquilla memory, including profile facts in USER.md and long-term notes in MEMORY.md or memory//.md.

opensquilla/opensquilla · 44 tokens

ha-data-stores

Map of Hope Agent's local data stores and safe read-only query workflow. Use when the user asks where Hope Agent stores data, wants to inspect sessions/messages/memory/logs/background jobs/knowledge indexes/settings, asks the model to query local app data, or debugging requires checking persisted state. Trigger…

shiwenwen/hope-agent · 115 tokens

establishing-project-context

Use when the user asks to establish shared project language, or project work exposes a conflicting, renamed, or deprecated domain term that needs active semantic modeling. Routine small tasks stay on the fast path.

GanyuanRan/Aegis · 45 tokens