not-goldfish: Instructions file for Claude Code

CLAUDE.md

not-goldfish CLAUDE.md is an instructions file for Claude Code from vmelooo/not-goldfish. It costs 812 tokens per session, scanned A, original, MIT.

A project instruction file for Claude Code and other AI coding assistants working on the not-goldfish repository. It documents the project's memory system, components, commands, invariants, and development conventions.

In plain words
What is it for?
Use it as repository-specific guidance when working on not-goldfish, including its SQLite store, command-line tool, daemon, hooks, integrations, and memory features.
Why use it?
It gives coding assistants shared context about how the repository is structured and how its tools are expected to behave.

Instructions file for Claude Code

Written for Claude Code: PreCompact hook event. Also seen: mentions CLAUDE.md; mentions Claude Code; mentions AGENTS.md.

This is vmelooo/not-goldfish's own configuration. It tells Claude Code how to work on not-goldfish itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything not-goldfish configures →

Reuse

Borrowing it

Nothing to install: this file belongs to vmelooo/not-goldfish. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/vmelooo/not-goldfish/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/vmelooo/not-goldfish

Made for: Claude Code.

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 not-goldfish CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/vmelooo/not-goldfish/claude-md/github.svg)](https://agentmods.dev/instructions/vmelooo/not-goldfish/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/vmelooo/not-goldfish/claude-md"><img src="https://agentmods.dev/badge/instructions/vmelooo/not-goldfish/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.

agentmods 80×15 button for not-goldfish CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/vmelooo/not-goldfish/claude-md"><img src="https://agentmods.dev/badge/instructions/vmelooo/not-goldfish/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 812 This file is loaded in full into every session.
When invoked 812 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00812 $0.00812
Opus 5.5 $0.00325 $0.00325
Sonnet 5.5 $0.00162 $0.00162
Haiku 4.5 $0.00081 $0.00081

Measured 2mo ago against content hash aca339530bae, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

not-goldfish 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 2mo 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 · 51 lines

How it starts

The opening of the file, as written. The whole thing — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.

not-goldfish

Memória universal, cache e higiene de contexto para harnesses de IA (Claude Code, Kimi Code, Gemini CLI, Codex). SQLite (WAL + FTS5) é a fonte de verdade; tudo mais (busca híbrida, grafo de sabedoria, UI web) é derivado dela.

Crates

Crate O que é
ng-core Store SQLite (WAL, FTS5), tags léxicas, embeddings (HashEmbedder + busca híbrida), grafo de sabedoria, paths, timeutil.
ng-hook Binário fino chamado pelos hooks do harness (<5ms), injeção proativa de memória, gate de higiene PreCompact.
ngd Daemon: writer único via socket Unix, worker de embedding/grafo em background, UI web (axum, thread própria).
ng-sessions Parsers tolerantes a versão dos transcripts de cada harness + rewrite seguro (backup + rename atômico).
ng-adapters Integração multi-harness pura/testável: instaladores de hooks, registro de MCP, sync de personas, dispatch, watcher de transcripts (Codex).
ng-cli CLI ng (install, search, status, daemon, ui, doctor, mcp, sync, dispatch, wisdom, memory, gain, sync-context).

Detalhes de uso, variáveis de ambiente e arquitetura completa: README.md.

Comandos

cargo build --release              # build de todo o workspace
cargo test --workspace --release   # todos os testes (rode -p <crate> para um só)
cargo fmt --all --check            # formatação (CI falha se não passar)
cargo clippy --workspace --all-targets -- -D warnings   # lints (CI trata warning como erro)
./target/release/ng doctor         # diagnóstico de ambiente instalado

SQLite é bundled (feature do rusqlite no Cargo.toml raiz) — compilado a partir de fontes vendored, sem apt install libsqlite3-dev nem nenhuma dependência de sistema.

Invariantes do projeto

  • Captura nunca quebra a sessão do harness. ng-hook sempre sai com código 0, mesmo em falha de captura — perder uma memória é aceitável, quebrar o fluxo do usuário não é.
  • Nada capturado é deletado. Eventos são o log de fonte de verdade; higiene/eviction só substitui conteúdo por um stub recuperável ([ng-evicted: ...]) ou remove do transcript ativo do harness — nunca do banco not-goldfish nem do backup em disco.
  • Hot path do hook tem orçamento de <5ms. ng-hook é o binário mais fino possível: parse do payload, tags léxicas, entrega ao daemon (ou fallback síncrono se o daemon estiver fora). Qualquer trabalho pesado (embeddings, grafo) roda em background em ngd, nunca no hook.
  • A UI web liga só em 127.0.0.1, nunca 0.0.0.0. Transcripts contêm prompts, saídas de ferramentas e às vezes segredos — não é uma superfície para expor na rede.

Read the full file on GitHub · 51 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. 2mo ago First seen · 51 lines · 812 tokens per session scan A aca339530bae

Subscribe to this mod's changes

not-goldfish CLAUDE.md is an instructions file published in the GitHub repository vmelooo/not-goldfish (1 stars, last pushed 3mo ago), licensed MIT. It adds 812 tokens to every session, about $0.0032 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 2026-08-31.

Related

Other instructions, from other repositories

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 9,715 tokens

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,461 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).

microsoft/vscode · 5,262 tokens

langchain AGENTS.md

AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain · 4,680 tokens

vscode copilot-instructions.md

Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).

microsoft/vscode · 2,750 tokens

deepseek-harness AGENTS.md

AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.

deepseek-ai/deepseek-harness · 4,261 tokens