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/human-turn/bcs-mcp/agents-mdgit clone --depth 1 https://github.com/human-turn/bcs-mcpWrote 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/human-turn/bcs-mcp/agents-md)<a href="https://agentmods.dev/instructions/human-turn/bcs-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/human-turn/bcs-mcp/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 | $0.03571 | $0.03571 |
| Opus 5 | $0.01785 | $0.01785 |
| Sonnet 5 | $0.00714 | $0.00714 |
| Haiku 4.5 | $0.00357 | $0.00357 |
Grade A, and why
bcs-mcp 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 3d 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 — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — инструкции для AI-агентов
MCP-сервер (TypeScript, stdio) для BCS Trade API (БКС Мир инвестиций). Агент, работающий с этим репо, обязан соблюдать инварианты безопасности — они важнее любых фич.
Сборка и проверка
npm install
npm run build # tsc, strict mode — должен проходить без ошибок
npm run dev # запуск из исходников (tsx)
Смоук-тест — JSON-RPC по stdio (без реального токена сервер стартует и отдаёт списки):
(printf '%s\n' \
'{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2025-06-18","capabilities":{},"clientInfo":{"name":"smoke","version":"0"}}}' \
'{"jsonrpc":"2.0","method":"notifications/initialized"}' \
'{"jsonrpc":"2.0","id":2,"method":"tools/list"}'; sleep 2) \
| BCS_REFRESH_TOKEN=fake node dist/index.js
Ожидаемо: 15 tools без флагов (14 read из read.ts + get_server_info);
с BCS_ALLOW_TRADING=true +3 (place/edit/cancel_order). Prompts: 10.
Реальные проверки — только с живым read-only токеном.
Структура
| Файл | Назначение |
|---|---|
src/index.ts |
Точка входа: регистрация tools/prompts/resources по env-флагам |
src/endpoints.ts |
Базовые URL сервисов be.broker.ru (единое место) |
src/client.ts |
config из env, TokenManager (Keycloak refresh→access), bcsFetch (retry 401/429) |
src/helpers.ts |
toMsk, mapPool, fail (HTTP-хинты), notifyProgress, computeChunks |
src/output.ts |
Файловые выгрузки outputPath: path-guard, CSV, summary |
src/instruments-cache.ts |
Мемо-кэш ticker→карточка (classCode, lotSize, MOEX-приоритет) |
src/catalog.ts |
Ленивый каталог тикеров для автодополнения prompt-аргументов |
src/prompts.ts |
Рецепты slash-команд: recipes + зонтичные композиции |
src/resources.ts |
MCP-ресурсы (эталон portfolio-target.json) |
src/tools/read.ts |
14 read-tools |
src/tools/trading.ts |
place/edit/cancel order + elicitation-подтверждение |
src/tools/info.ts |
Диагностика для фидбэков |
Инварианты безопасности (НЕ ослаблять)
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.
- 3d ago First seen · 163 lines · 3,571 tokens per session scan A 1bfaadbe5d40
bcs-mcp AGENTS.md is an instructions file published in the GitHub repository human-turn/bcs-mcp (2 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 3,571 tokens to every session, about $0.0179 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
Convertible-Bond-Pricing-Research CLAUDE.md
Instructions for ericxuzhesheng/Convertible-Bond-Pricing-Research, covering claude.md — convertible bond pricing research, current published state and routine commands.
bukio-cli AGENTS.md
AGENTS.md instructions for erikvankempen/bukio-cli, covering agents.md — bukio-cli agent manual, 1. house rules (non-negotiable), 2. environment, 3. command quick reference and 3.1 jurisdiction profiles (16 markets).
bullet-bots AGENTS.md
Instructions for bulletxyz/bullet-bots, covering bullet-bots, build & test, run the faucet curl printed by keygen, then, or (base58 from phantom/delegation export, or hex) and architecture — the harness, feeds, and actors.
romaco-mcp AGENTS.md
AGENTS.md instructions for romaco-labs/romaco-mcp, covering purpose, read first, product boundaries, architecture invariants and development method.
financial-services-plugins-new CLAUDE.md
Claude Code instructions for yuping322/financial-services-plugins-new, covering financial services plugins, repository structure, plugin structure, key files and development workflow.
zenmoney-mcp CLAUDE.md
Claude Code instructions for sakost/zenmoney-mcp, covering zenmoney mcp server — development guidelines, project overview, build & test, architecture and coding standards.