Borrowing it
Nothing to install: this file belongs to isamu/ever-better. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/isamu/ever-better/main/CLAUDE.mdgit clone --depth 1 https://github.com/isamu/ever-betterWrote 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.
[](https://agentmods.dev/instructions/isamu/ever-better/claude-md)<a href="https://agentmods.dev/instructions/isamu/ever-better/claude-md"><img src="https://agentmods.dev/badge/instructions/isamu/ever-better/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/isamu/ever-better/claude-md"><img src="https://agentmods.dev/badge/instructions/isamu/ever-better/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.03220 | $0.03220 |
| Opus 5.5 | $0.01288 | $0.01288 |
| Sonnet 5.5 | $0.00644 | $0.00644 |
| Haiku 4.5 | $0.00322 | $0.00322 |
Grade A, and why
ever-better CLAUDE.md 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.
How it starts
The opening of the file, as written. The whole thing — 228 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — ever-better
Working notes for AI agents in this repo. What the tool is and how to use it lives in
README.md; the design reasoning lives in plans/mvp-p0-p2.md; where the
work stands and what to do next lives in docs/HANDOFF.md. Read those first.
Stack
TypeScript, ESM, Node 20.11+. Zero runtime dependencies — that is a feature, not an accident.
A tool that installs other people's dev dependencies has no business dragging in its own. Before
adding one, check whether node:util, node:fs or a subprocess already covers it.
Package manager here is yarn.
Run after changes
yarn format # prettier
yarn lint # eslint (this repo runs every tier it generates)
yarn typecheck # tsc --noEmit over src AND test
yarn build # tsc -p tsconfig.build.json -> dist/
yarn test # node:test over test/*.ts
Never judge these through a pipe — yarn lint | tail exits with tail's status and reports a
failing lint as success. Check the exit code, or use set -o pipefail.
Imports use .ts extensions
import { x } from "./foo.ts", not "./foo.js". Tests run the TypeScript sources directly under
Node's native type stripping, and rewriteRelativeImportExtensions turns them into .js on emit.
erasableSyntaxOnly is on, so enums, namespaces and parameter properties are compile errors — they
would build fine and then crash the test runner.
The one architectural rule
Deterministic work goes in the CLI. Judgment goes in a skill.
CLI (src/) |
Skill (skills/) |
|---|---|
| detect, install, count, render, gate | is this warning a real bug |
| the same answer every run | how to split this function |
| no LLM in the loop | DRY or coincidence |
If you are about to write a heuristic in a SKILL.md that produces a number, it belongs in the CLI.
If you are about to encode taste in src/, it belongs in a skill.
Pure decisions, one impure gatherer
gatherFacts(cwd) in src/facts.ts is the only place detection touches the filesystem. It
returns a RepoFacts value; diagnose(facts) and everything under src/detect/ are pure
functions over it. That is why the diagnosis has real tests and no fixtures on disk.
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.
- 6d ago First seen · 228 lines · 3,220 tokens per session scan A dbd69726bac9
ever-better CLAUDE.md is an instructions file published in the GitHub repository isamu/ever-better (28 stars, last pushed 2d ago), licensed MIT. It adds 3,220 tokens to every session, about $0.0129 per session on Opus 5.5. 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 2026-11-10.
ever-better CLAUDE.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| ever-better CLAUDE.mdisamu/ever-better | Instructions file | 28 | 2d ago | MIT | A | 3,220 | Claude Code |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 144k | today | MIT | A | 7,461 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 5,685 | GitHub Copilot |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 150k | today | MIT | A | 4,680 | Codex, OpenCode |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 3,086 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 282k | 3d ago | MIT | A | 4,335 | Codex, OpenCode |