Improve System

Mine your own Claude Code history for what keeps going wrong, then fix it at the source. Finds corrections you had to give more than once (a rule should have stopped them), jobs you keep typing by hand (a missing skill), context you keep pasting in (a missing fact), and the kinds of work eating the most hours. Use when the user says "improve the system", "improve my setup", "what am I correcting", "what keeps going wrong", "mine my history", "what skills should I build", "why do I keep repeating myself", or once or twice a week as a routine. Also use with a topic to audit one kind of work that

10 stars 1 forksAI, LLMs & Agents Plugin marketplace

Install

Claude Code plugin
/plugin marketplace add charlie947/skills-for-everyone

The repo is a Claude Code plugin marketplace. Run /plugin afterwards to pick the plugin that includes this skill.

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add charlie947/skills-for-everyone --skill improve-system
Manual install
git clone --depth 1 https://github.com/charlie947/skills-for-everyone.git /tmp/skills-for-everyone
mkdir -p ~/.claude/skills
cp -r /tmp/skills-for-everyone/skills/improve-system ~/.claude/skills/improve-system

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# improve-system

You correct Claude, it improves for one chat, and next week you give the same correction again. Your own history already records every one of those moments. This skill reads it, finds what keeps coming back, and fixes it in the right place so you stop repeating yourself.

## Step 1: Run the miner

The miner is a small Python script that ships with this skill. It reads the prompts you typed into Claude Code (`~/.claude/history.jsonl`). It changes nothing.

Run the miner from this skill's folder:

```bash
python3 "${CLAUDE_SKILL_DIR}/scripts/improve-system.py" --days 7
python3 "${CLAUDE_SKILL_DIR}/scripts/improve-system.py" --days 7 --topic 'email|inbox|reply'
```

`${CLAUDE_SKILL_DIR}` is the folder this skill is installed in, so the command works for a plugin install and a copied folder alike. If it comes up empty (outside Claude Code), use the path of the folder this SKILL.md sits in.

The first is the whole setup. The second narrows it to one kind of work. It finishes in a second or two and writes a dated report, plus a `.json` file with every correction quote, to `~/.claude/logs/improve-system/`. Read the JSON, not just the summary.

If it says there is no history, the user has not used Claude Code on this computer yet. Stop and tell them to come back after a week of normal use.

The correction detector is a phrase list. It catches most pushback and some false alarms. Treat each quote as a claim, and drop the ones that are not really a correction.
Read the full skill on GitHub

Details

Skill name
improve-system
Path
skills/improve-system/SKILL.md
License
MIT
Repo updated
6 October 2026

FAQ

What does the Improve System skill do?

Mine your own Claude Code history for what keeps going wrong, then fix it at the source. Finds corrections you had to give more than once (a rule should have stopped them), jobs you keep typing by hand (a missing skill), context you keep pasting in (a missing fact), and the kinds of work eating the most hours. Use when the user says "improve the system", "improve my setup", "what am I correcting", "what keeps going wrong", "mine my history", "what skills should I build", "why do I keep repeating myself", or once or twice a week as a routine. Also use with a topic to audit one kind of work that

How do I install the Improve System skill?

Claude Code plugin: /plugin marketplace add charlie947/skills-for-everyone. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Improve System popular?

It lives in charlie947/skills-for-everyone, which has 10 GitHub stars and 1 forks.

Related skills

Answer First

10

charlie947/skills-for-everyone

Shape every answer so the bottom line comes first. Sentence one is the answer, the number, the verdict or the one thing to do. Everything after it is support. Use when the user asks for output that is short, skimmable, action-led or "no waffle", and keep applying it for the whole conversation once invoked, not just one reply. Triggers on "answer first", "bottom line first", "no preamble", "no waffle", "keep it short", "just tell me what to do", "cut the fluff", "ADHD mode", or any request to stop burying the answer in paragraph four. It shapes Claude's replies to you, not writing you publish.

Writing & Content

Close

10

charlie947/skills-for-everyone

End a good session properly. Banks what the session taught into Claude's memory so the next session starts smarter, finishes every loose end it can, proves the changes actually landed, and hands back only what genuinely needs you. Four phases, in order: LEARN, FIX, PROVE, REPORT. Use when the user says "close", "/close", "close out", "wrap this up", "are we good to close?", "what's still open?", "anything left?", "did we miss anything", "save progress", "bank what we learned", "remember this for next time", "lock in our corrections", or otherwise signals a session is ending. Not for handing wo

Learning & Education

Prompt Lab

10

charlie947/skills-for-everyone

Turn an idea for a prompt into a finished, tested prompt in one turn. You type the idea, and the prompt comes back in a code block, graded out of 100 against Anthropic's current published prompting guidance (fetched live every run) and a plain bar any reader can run. The pass mark is 95, and Claude runs the prompt itself before handing it over whenever it can. Use when the user wants a prompt written, improved or checked. Triggers on "prompt lab", "write me a prompt", "make this prompt better", "grade this prompt", "score this prompt", "is this prompt any good", "make this prompt expert level"

AI, LLMs & Agents

Anti AI Slop Design

9

prangishviliAbe/agent-skills

Design, refine, implement, or critique websites, landing pages, dashboards, apps, and components so nothing looks AI-generated or templated. Use for "looks AI", "too generic", "make it premium", "make it look better", art direction, redesigns, visual polish, a page or component designed from scratch, or a review of how a UI looks, including Georgian and multilingual layouts. Covers every layer where slop appears: page structure, typography, color, spacing, surfaces, icons, imagery, copy, motion, and the code that produces them. Replaces default choices with decisions derived from the subject,

AI, LLMs & Agents

Short Context Agent

5

che-shr-cat/short-context-agent

Keep agentic work inside a short context window using the NVIDIA "You don't need long context" recipe (arXiv:2611.05306) — forward-looking handoff briefs instead of generic summaries, End-of-Conversation / Single-Action compaction, and selective reading of large inputs with grep + bounded windows instead of pasting whole files into the prompt. Use this skill whenever a task is long-horizon (multi-step debugging, migrations, multi-file refactors, long investigations or research), whenever the user mentions context limits, compaction, /compact, handoff, checkpointing, "resume where we left off",

AI, LLMs & Agents

Adobe Express Core

9

Sandgrouse/adobe-express-dev-skill

Adobe Express environment essentials, two-runtime architecture, project setup, and MCP server configuration. Use when starting add-on development, understanding iframe vs sandbox boundaries, configuring MCP servers, setting up local projects, or validating manifest configuration.

AI, LLMs & Agents

Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.