skill CX Codex

continual-learning · cursor — Claude Code skill: what it does, how to install

Q: How do I install continual-learning in Claude Code?

Run `npx skills add cursor/plugins --skill continual-learning -a claude-code`. The skill then lives at `~/.claude/skills/continual-learning/SKILL.md`.

Q: What does continual-learning do?

It drives the memory update loop: calls the `agents-memory-updater` subagent and returns its result. The parent skill does not read transcripts or edit files.

A skill that runs the continual-learning loop: distill past chats into AGENTS.md, with all the actual work handed to the agents-memory-updater subagent.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓

npx skills add cursor/plugins --skill continual-learning -a claude-code

Open on GitHub

What continual-learning does

The skill keeps AGENTS.md current while doing none of the mining itself. Its job is orchestration: call the agents-memory-updater subagent, wait, return whatever that subagent produced. Reading transcripts and editing files stays over there.

Two steps, and that is the whole flow. The guardrails are strict. The parent skill must not mine chats or touch files on its own, and it must not skip the subagent and do the work directly. So if agents-memory-updater is not in the project, the chain has nothing to run — install that one first.

One more line in the frontmatter matters: disable-model-invocation: true. The model will not pick this skill up unprompted, you have to ask for it by name. Worth writing that trigger into your project instructions, or the file sits unused at .cursor/skills/continual-learning/SKILL.md on Cursor and at the equivalent path elsewhere.

What it helps with

turn conclusions from earlier chats into entries in AGENTS.md
refresh project rules without rereading old logs
trigger the memory subagent with one request
keep orchestration and file edits in separate agents

Example prompts for the agent

Run continual-learning: mine the last few chats and refresh AGENTS.md.
Use the continual-learning skill and return the output from agents-memory-updater.
Keep AGENTS.md up to date — trigger continual-learning instead of editing it yourself.

How to install continual-learning

1Skills CLI (Vercel)

npx skills add cursor/plugins --skill continual-learning -a claude-code

2Manually: put the files in place

~/.claude/skills/continual-learning/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/continual-learning/

1Skills CLI (Vercel)

npx skills add cursor/plugins --skill continual-learning -a codex

2Manually: put the files in place

.agents/skills/continual-learning/SKILL.md

For all projects use ~/.agents/skills/

1Skills CLI (Vercel)

npx skills add cursor/plugins --skill continual-learning -a cursor

2Manually: put the files in place

.cursor/skills/continual-learning/SKILL.md

Cursor also reads .agents/skills/ and .claude/skills/

1GitHub CLI

gh skill install cursor/plugins continual-learning

2Manually: put the files in place

.github/skills/continual-learning/SKILL.md

1Skills CLI (Vercel)

npx skills add cursor/plugins --skill continual-learning -a gemini-cli

2Platform CLI

gemini skills install https://github.com/cursor/plugins --path continual-learning/skills/continual-learning

Manually: put the files in place

.devin/skills/continual-learning/SKILL.md

The legacy .windsurf/skills/ path still works

Skills CLI (Vercel)

npx skills add cursor/plugins --skill continual-learning -a opencode

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about continual-learning

How do I install continual-learning in Claude Code?

Run npx skills add cursor/plugins --skill continual-learning -a claude-code. The skill then lives at ~/.claude/skills/continual-learning/SKILL.md.

What does continual-learning do?

It drives the memory update loop: calls the agents-memory-updater subagent and returns its result. The parent skill does not read transcripts or edit files.

Do I need the agents-memory-updater subagent?

Yes. The skill's own rules forbid bypassing it, so without that subagent there is nothing to delegate to. On top of that, disable-model-invocation: true means you invoke the skill explicitly.

continual-learning by other authors comes first.

continual-learning microsoft A skill that teaches AI agents to remember mistakes, tool patterns, and project conventions between sessions. Without it, each run starts fresh; with it, agents compound their knowledge and work more accurately. 79installs self-learning kulaxyz A meta-skill that captures a working route from the current session as a reusable skill, so the next session starts already knowing it. CX Codex +1 850installs groove-utilities-memory-graduate andreadellacorte Promotes a stable workflow insight from .groove/memory/learned/ into AGENTS.md, so the rule applies to every future session without running prime. CX Codex +1 510installs claude-migrate-memory-to-doc daymade Pulls your profile and preferences out of Claude's personal memory into documents that Claude, Cursor and other CLIs can read. CX Codex +1 325installs session-learning-capture davila7 A slash command that pulls lessons out of the current Claude session and writes them into the project's memory files, so the next run starts informed. 425installs context-engineering-collection muratcankoylan Patterns for production agent design: context management, multi-agent systems, long-running loops. Use when building autonomous systems that need to operate reliably. CX Codex +1 4,200installs ai-agents-architect sickn33 A skill that breaks an agent into parts: ReAct loop, planning, memory layers, multi-agent orchestration. Useful once your agent exists and you need to decide how it fails. CX Codex +1 1,500installs

From the same set45

get-pr-comments Pulls feedback and discussions from the active pull request. It sorts reviewer remarks by importance so you can jump straight into fixes. CX Codex +1 2,200installs workflow-from-chats Mines recent agent conversations to turn recurring feedback and habits into skills, rules, or workflow guides. CX Codex +1 2,100installs thermos Runs two code-review subagents in parallel as background tasks, then folds their findings into one prioritised report. CX Codex +1 2,100installs run-smoke-tests This skill runs Playwright smoke tests, digs into failures and confirms a fix holds. Handy before a release or after a risky change. CX Codex +1 2,100installs cli-for-agents Designs and audits terminal tools so coding agents can run commands headlessly without getting stuck in interactive menus. CX Codex +1 1,400installs review-plugin-submission Audits a Cursor plugin for marketplace readiness: manifest, component metadata, discovery paths, docs quality. CX Codex +1 1,200installs create-plugin-scaffold Builds a Cursor plugin scaffold: manifest, component folders and a marketplace entry. Handy when a plugin starts from an empty folder. CX Codex +1 1,200installs docs-canvas Turns architecture notes, API specs, and walkthroughs into an interactive Cursor Canvas with navigation and diagram cards. CX Codex +1 1,200installs