Skip to content
Agentmart
Submit a tool

memory

Memory — an architecture concept covering memory, tier, promotion, lifecycle, consolidation.

Skill by mmc7676

IndexedApache-2.0
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/mmc7676-memory

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Memory — an architecture concept covering memory, tier, promotion, lifecycle, consolidation. Use when the user's task concerns memory, or matches: memory, tier, promotion, lifecycle, consolidation, episodic, working, persistent, semantic. Route with loops_route first — this concept's 8 skills and 14 loops are listed in reference/, and a coherent hit often names a different concept's formation.

Use cases: productivity

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/memory

What's inside

Skill name
memory
Description
Memory — an architecture concept covering memory, tier, promotion, lifecycle, consolidation. Use when the user's task concerns memory, or matches: memory, tier, promotion, lifecycle, consolidation, episodic, working, persistent, semantic. Route with loops_route first — this concept's 8 skills and 14 loops are listed in reference/, and a coherent hit often names a different concept's formation.
License
Apache-2.0
Instructions
33 lines in SKILL.md · view on GitHub

Files (4)

  • SKILL.md
  • agents.md · reference/
  • loops.md · reference/
  • skills.md · reference/

SKILL.md (first lines)

# Memory

**Architecture concept** · 8 skill(s) · 14 loop(s) tagged `memory`

Vocabulary: `memory`, `tier`, `promotion`, `lifecycle`, `consolidation`, `episodic`, `working`, `persistent`, `semantic`

## When this applies

A task concerns `memory` when it matches that vocabulary — but matching the vocabulary is not the same as belonging here. Route first and let the decision be examinable.

## What to do

1. **Route.** Call `loops_route(task="...")` with the task in plain language. It scores the task across concepts, skills and agents and returns the evidence for its decision — what it considered, what it rejected, and why. It binds nothing.
2. **Browse what already exists.** `loops_catalog(concept="memory")` returns the same loops with their live status. Prefer one of them to anything hand-built.
3. **Bind, if you are going to run it.** `context_route_bind` asks the same question and commits: it binds the matching loop's partition in this session's own window. Without the Context plane installed, routing still answers; it just cannot be followed by a bind.

## What ships with this skill

| file | contents |
|---|---|
| [reference/skills.md](reference/skills.md) | the 8 skill(s) under this concept, each with the agent formation bound to it |
| [reference/loops.md](reference/loops.md) | the 14 loop(s) tagged `memory`, with provenance, status and hardening record |
| [reference/agents.md](reference/agents.md) | the registry agents that appear in those loops, and what each specializes in |

14 loop(s) in the library are tagged `memory` (14 validated). They are listed with their hardening records in [reference/loops.md](reference/loops.md).

## The rule that carries

A refusal is a result. If the runtime refuses a read, report it with its own message and hint rather than routing around it — the hint names the bridge that would make the access legal, and whether that access is justified is a decision to make explicitly.

See [skills/scw](../scw/SKILL.md) for how partitioning and routing actually work.

From mmc7676/Maxey0, Apache-2.0.

More from mmc7676/Maxey0

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Sep 25, 2026 (1 month ago)
Last release
Sep 28, 2026 (1 month ago)

From GitHub (fetched Oct 24, 2026).

Details

License
Apache-2.0
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from mmc7676

Cite this page

memory skill. Agent Mart. https://agentmart.dev/skills/mmc7676-memory (last updated 2026-10-24)