statuslin.es CLAUDE.md

statuslin.es CLAUDE.md is an instructions file for coding agents from NathanAB/statuslin.es. It costs 2,870 tokens per session, scanned A, original, MIT.

A project guide for statuslin.es, a website where people browse and copy Claude Code status-line configurations. It describes the application’s architecture, technology choices, and required quality checks.

In plain words
What is it for?
Use it when changing the submission flow, sandbox rendering, review queue, gallery, authentication, database code, or status-line adoption pages.
Why use it?
It helps coding agents follow the project’s boundaries and run its required linting, type checks, tests, and other safeguards.

Instructions file

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.

agentmods
npx agentmods add instructions/nathanab/statuslin.es/claude-md
Clone the repo
git clone --depth 1 https://github.com/NathanAB/statuslin.es

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 statuslin.es CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/nathanab/statuslin.es/claude-md.svg)](https://agentmods.dev/instructions/nathanab/statuslin.es/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/nathanab/statuslin.es/claude-md"><img src="https://agentmods.dev/badge/instructions/nathanab/statuslin.es/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 2,870 This file is loaded in full into every session.
When invoked 2,870 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
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 $0.02870 $0.02870
Opus 5 $0.01435 $0.01435
Sonnet 5 $0.00574 $0.00574
Haiku 4.5 $0.00287 $0.00287

Measured 4d ago against content hash e9b10294eb51, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

statuslin.es 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 4d 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.

CLAUDE.md · 84 lines

How it starts

The opening of the file, as written. The whole thing — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.

statuslin.es

Community gallery of Claude Code status lines — browse configs as rendered previews and copy one to use.

This is an agent-first codebase. Most changes are written by AI agents. The standards below are enforced mechanically (linter, typecheck, tests, git hooks, CI) and by every agent on itself. Do not bypass the gates — no git commit --no-verify, no skipping tests, no claiming "done" without showing green output in the same message.

Stack

Bun (runtime + toolchain) · Vite · TanStack Start (React, SSR) · Better Auth (GitHub) · Drizzle + Postgres · E2B (untrusted-script sandbox). Dependencies are pinned exact — no ^ ranges (TanStack Start is RC, Nitro is beta). Upgrades are deliberate, smoke-tested events, never incidental.

Architecture

The lifecycle of a config: submit (src/submit) → a render job is queued → the worker renders it in the E2B sandbox (src/render) → it lands in the review queue (src/review) → an admin publishes → it appears in the gallery (src/gallery), where visitors copy it to use (src/adopt).

src/ map (one responsibility each):

  • routes/ — TanStack Start file-based routes: pages + API handlers
  • submit/ — submission form + flow (slug, obfuscation checks, allowed network hosts)
  • render/ — render pipeline: real E2B runner, fake runner (tests/no key), ANSI parsing, scenarios
  • review/ — admin review queue and publish/reject decisions
  • gallery/ — gallery list queries
  • votes/ — retained legacy voting code · adopt/ — copy/install a config
  • og/ — Open Graph card images · legal/ — terms page
  • db/ — Drizzle schema + migration client · lib/ — shared utils (env.ts)
  • ui/ — closed design-system components · styles/ — tokens (app.css)
  • server/ — Nitro server plugins (error context, PostHog)

The quality bar — non-negotiable

  1. TDD. Red → green → refactor for all new behavior. Write the failing test first and confirm it fails for the right reason. Deviations are allowed only for pure config, pure type definitions, or mechanical refactors covered by existing tests — and you must state the category out loud before deviating.
  2. Run the gate before every commit, show the evidence. bun run check (typecheck + lint + test) must be green, and the output must appear in the same message as any "done / passing / fixed" claim. A run from three messages ago does not count.
  3. Single source of truth (DRY). No magic strings or numbers duplicated across files. Environment-specific config — URLs, ports, secrets — lives in exactly one place (env vars); everything else derives from it. (Incident 2026-05-29: localhost:3000 was hardcoded in four files; now BETTER_AUTH_URL is the only source and the dev port + auth origins derive from it.)
  4. YAGNI. Build only what the spec/plan specifies. No speculative features, flags, or abstractions.
  5. Small, focused files. One clear responsibility per file; split when a file starts doing too much.
  6. Clear names. Name things for what they do, not how they work.
  7. Untrusted by default. Submitted status line scripts are hostile until proven otherwise — the E2B sandbox is the safety boundary; trust comes from the supply-chain controls (open-source + human review + hash-pinned immutable versions + re-review on every update). See SECURITY.md.

Read the full file on GitHub · 84 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. 4d ago First seen · 84 lines · 2,870 tokens per session scan A e9b10294eb51

Subscribe to this mod's changes

statuslin.es CLAUDE.md is an instructions file published in the GitHub repository NathanAB/statuslin.es (10 stars, last pushed 16d ago), licensed MIT. It adds 2,870 tokens to every session, about $0.0144 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 2026-08-31.