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.
npx agentmods add instructions/andrewcigan/vibe-dev-plugin/claude-mdgit clone --depth 1 https://github.com/andrewcigan/vibe-dev-pluginWrote 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/andrewcigan/vibe-dev-plugin/claude-md)<a href="https://agentmods.dev/instructions/andrewcigan/vibe-dev-plugin/claude-md"><img src="https://agentmods.dev/badge/instructions/andrewcigan/vibe-dev-plugin/claude-md.svg" alt="Measured on agentmods" 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 | $0.04392 | $0.04392 |
| Opus 5 | $0.02196 | $0.02196 |
| Sonnet 5 | $0.00878 | $0.00878 |
| Haiku 4.5 | $0.00439 | $0.00439 |
Grade A, and why
vibe-dev-plugin 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.
How it starts
The opening of the file, as written. The whole thing — 167 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Vibe Dev v8 — Мозг плагина
✅ v8 — провенанс + управляемый контекст поверх enforcement-фундамента. Механизмы (актуальное число и живой статус — только в docs/traceability.md, сейчас 67 отслеживаемых; экранные детекторы — язык-ловец
clarity-detectorи маска секретов в выводе — честно помечены display-only/частичными и НЕ считаются enforcement): провенанс фич как event-sourcing (горячая проекция + append-only лог + архив по ссылке; crash-safe записьrecord-change; git pre-commit гейты когерентности/append-only/архива без загрузки тела в контекст), стадия детализации M/L-фичи (не входит в active безdocs/changes/<id>/proposal.mdс приоритизированной P1 user story в Given/When/Then; backlog ленивый), управляемый/checkpointвместо рулетки авто-сжатия (cold-start gate блокирует шаблонный/некогерентный старт) + трёхуровневая модель контекста + сужение возврата читающих агентов, контракт фронтматтера агентов (пины моделей по стадиям +effort/disallowedTools; adversarial fresh-context verifier физически не пишет код), evidence на logic-фиче (runtime/e2e, не только typecheck) + negative-gate на M/L, бюджеты tool-call, единая цифра готовности в/audit. Фундамент v6.2/v7 держится: активация хуков доказуема (heartbeat + двухфазный профиль pending→strict + независимый git pre-commit backstop +/doctor), fail-loud (краш сторожа не молчит), clarity-gate на финальное сообщение (Stop-block с precision-гейтом на корпусе боевых сессий), evidence по поверхности фичи (монотонная строгость), обязательный research перед архитектурой (lock-паттерн), closing-mode, секрет-гигиена, config-protect, interrupt-recovery. Построено по решениям владельца (карточки c1-c12) + донорам pilotfish/OpenSpec/spec-kit/agents-best-practices + аудиту боевых сессий + независимой критике. id плагина —vibe-dev(версия 8.0.0, публичный github andrewcigan/vibe-dev-plugin).
Идентичность
Ты — Vibe Dev v8, harness-first система разработки продуктов для предпринимателя-непрограммиста. Берёшь бизнес-идею и автономно превращаешь в работающий продукт. Все технические решения принимаешь сам.
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.
- 4d ago First seen · 167 lines · 4,392 tokens per session scan A dba2c87443d4
vibe-dev-plugin CLAUDE.md is an instructions file published in the GitHub repository andrewcigan/vibe-dev-plugin (5 stars, last pushed 1mo ago), licensed MIT. It adds 4,392 tokens to every session, about $0.0220 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.
Other instructions, from other repositories
codebay CLAUDE.md
Claude Code instructions for khromov/codebay, covering claude.md, what this is, runtime & commands, comments and architecture.
mumei AGENTS.md
Instructions for iroha924/mumei, covering agent reviewer guidance, what mumei is, setup commands, distribution boundary — flag violations and code style.
mumei CLAUDE.md
Instructions for iroha924/mumei, covering mumei development guide, what mumei is, language policy, bash + jq conventions (essentials) and schemas/ conventions (essentials).
vibe-engineering CLAUDE.md
Instructions for ash1794/vibe-engineering, covering claude.md, what this repo is, plugin structure invariants — do not break these, 2. the plugin directory must not be the repo root and 3. marketplace name must differ from plugin name.
vibe-engineering AGENTS.md
Instructions for ash1794/vibe-engineering, covering vibe-engineering, skills, invocation, tool mapping and project conventions.
dhpk CLAUDE.md
Instructions for hmj1026/dhpk, covering gitnexus — code intelligence, always do, never do, resources and cli.