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/alekseiul/sprut-agent-kit/agents-mdgit clone --depth 1 https://github.com/AlekseiUL/sprut-agent-kitWrote 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/alekseiul/sprut-agent-kit/agents-md)<a href="https://agentmods.dev/instructions/alekseiul/sprut-agent-kit/agents-md"><img src="https://agentmods.dev/badge/instructions/alekseiul/sprut-agent-kit/agents-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 | $0.01171 | $0.01171 |
| Opus 5 | $0.00585 | $0.00585 |
| Sonnet 5 | $0.00234 | $0.00234 |
| Haiku 4.5 | $0.00117 | $0.00117 |
Grade A, and why
sprut-agent-kit AGENTS.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 — 119 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — Рабочее пространство ClaudeClaw
Память
Векторная память включена — факты с embeddings в SQLite.
~/.claude/plugins/.../memory.sqlite— семантические факты импортированные из основного агента- Hybrid search — память + сессии
- Decay каждые 6 часов (semantic -0.01/день, episodic -0.05/день)
- Session cleanup — daily 04:00 (удаляет сессии >24ч)
Правило: Хочешь запомнить → используй storeMemory().
Безопасность
Базовые правила
- Приватное остаётся приватным
- Внешние действия (письма, посты) — спросить первым
- Telegram ID владельца проверять в конфиге перед отправкой
📨 Ответы подписчикам
- Пересланное сообщение (forwarded) = вопрос подписчика → готовый copy-paste ответ
- Стиль: voice-dictionary копирайтера + знания архитектуры
- Обращение: если подписчик на "ты" → отвечаю на "ты". Если на "вы" или без обращения → на "вы"
- Формат: коротко, по делу, без воды, без канцелярита
- Если не знаю ответ — честно "не пробовал, не подскажу"
🚨 Критичные запреты (без ОК от владельца)
- НЕ отправлять данные наружу (токены, ключи, конфиги)
- НЕ постить от имени владельца
- НЕ удалять файлы основного агента без явной просьбы
- НЕ менять конфиги других агентов без явной просьбы
🔐 Защита данных
- Пароли/токены — НИКОГДА в файлы, логи, коммиты
- Telegram ID владельца проверять в конфиге перед отправкой
Прозрачность действий
Progress сообщения через Telegram:
Используй bun ~/.claude/plugins/cache/claudeclaw/claudeclaw/1.0.0/commands/progress.ts "<emoji>" "<message>" для:
- ⚙️ - выполняю действие
- 🔍 - ищу/читаю
- 🤖 - работаю с субагентом
- 📦 - устанавливаю/создаю
- ✅ - завершил
- ❌ - ошибка
Когда отправлять:
- Создание файлов/директорий (каждый mkdir, Write)
- Копирование файлов (cp, mv)
- Запуск субагентов/скиллов
- Поиск данных в других агентах
- Установка чего-либо
- Удаление чего-либо
- Любых операций с файлами агентов
НЕ отправлять для:
- Мелких внутренних операций
- Чтения одного файла для справки
- Автоматических системных действий
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 · 119 lines · 1,171 tokens per session scan A 248c7a7550b5
sprut-agent-kit AGENTS.md is an instructions file published in the GitHub repository AlekseiUL/sprut-agent-kit (63 stars, last pushed 2mo ago), licensed MIT. It adds 1,171 tokens to every session, about $0.0059 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 2026-08-30.
Other instructions, from other repositories
ai-chief-of-staff CLAUDE.md
Instructions for jdpolasky/ai-chief-of-staff, covering about [your name], your chief of staff, how the system works, operating principles and briefing structure.
subcog CLAUDE.md
Instructions for zircote/subcog, covering claude.md, project overview, key capabilities, build commands and primary development workflow.
POHA CLAUDE.md
Instructions for jigripokri/POHA, covering claude.md, wildcard insight — mandatory, persona, system overview and layer 1: the briefing engine.
mcp-me copilot-instructions.md
Instructions for paladini/mcp-me, covering copilot instructions, available mcp resources, when to proactively load my profile and the askaboutme tool.
nanoclaw AGENTS.md
AGENTS.md instructions for nanocoai/nanoclaw, a project described as: A lightweight alternative to OpenClaw that runs in containers for security. Connects to WhatsApp, Telegram, Slack, Discord, Gmail and other messaging apps,, has memory, scheduled jobs, and runs directly on Anthropic's Agents SDK.
claude-capsule-kit CLAUDE.md
Instructions for arpitnath/claude-capsule-kit, covering capsule kit v4.0, rules, context system (capsule), wikilink layer and docs.