troitsa

troitsa is a skill for Claude Code, Codex from HinkoK/agent-skills. It costs 88 tokens per session (2,202 once invoked), scanned A, original, MIT.

A workflow that gives one task three separate roles: a planner, a worker that handles the main workload, and a critic that tries to find problems. The roles may be assigned to different AI models or acted out within one model.

In plain words
What is it for?
Use it for strategy, research summaries, product choices, business decisions, content direction, and other work where quality matters more than speed.
Why use it?
It separates planning, production, and checking so important decisions and research are less likely to rely on one unchallenged answer.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions Gemini CLI.

Good fit Use it for strategy, research summaries, product choices, business decisions, content direction, and other work where quality matters more than speed.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/hinkok/agent-skills/troitsa
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.

Any agent
npx skills add HinkoK/agent-skills --skill troitsa
Clone the repo
git clone --depth 1 https://github.com/HinkoK/agent-skills

Made for: Claude Code, Codex.

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 troitsa

README.md
[![agentmods](https://agentmods.dev/badge/skills/hinkok/agent-skills/troitsa.svg)](https://agentmods.dev/skills/hinkok/agent-skills/troitsa)
Your own site
<a href="https://agentmods.dev/skills/hinkok/agent-skills/troitsa"><img src="https://agentmods.dev/badge/skills/hinkok/agent-skills/troitsa.svg" alt="Measured on agentmods" height="20"></a>
Per session 88 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,202 The whole file, excluding the scripts and references it only reads on demand.
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 $0.00088 $0.02202
Opus 5 $0.00044 $0.01101
Sonnet 5 $0.00018 $0.00440
Haiku 4.5 $0.00009 $0.00220

Measured 7d ago against content hash 8f336b2575c8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

troitsa 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 7d 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.

skills/troitsa/SKILL.md · 163 lines

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.

Троица

Режим, в котором одна задача проходит через три модели с разными ролями. Смысл не в «трёх моделях ради красоты», а в разделении труда: дорогая умная модель только думает и собирает итог, дешёвая делает весь объём, а отдельная независимая модель пытается результат сломать. Так выходит и дешевле, и качественнее, чем одной моделью.

Hermes usage rules

  • Use only for non-trivial tasks where quality matters more than speed: strategy, content direction, product decisions, research synthesis, business/niche choices, and high-stakes planning.
  • Do not claim real model calls were made unless they were actually made through available tools/providers.
  • If separate Worker/Critic models are unavailable, run it as a role workflow inside the current agent and say that the critic is not independent.
  • Keep the final answer concise: show internal stages only if the user asks.

Три модели и их роли

Рекомендуемая конфигурация по умолчанию:

GPT      = Дирижёр (Conductor) — думает, планирует, ставит ТЗ, собирает финал. Включается мало → стоит мало.
DeepSeek = Работяга (Worker)   — делает основной объём: анализ, варианты, черновики, ресёрч. Дешёвый, его не жалко гонять.
Gemini   = Критик  (Critic)    — ломает результат работяги, ищет дыры. Независимый и часто дешёвый/бесплатный.
Hermes   = система, которая связывает всё вместе и возвращает пользователю чистый финал.

Можно заменить конкретные модели, но нельзя ломать разделение ролей:

  • Дирижёр / Conductor — самая сильная доступная модель для рассуждения, постановки ТЗ и финальной сборки.
  • Работяга / Worker — дешёвая high-throughput модель для объёма, перебора вариантов, черновиков и ресёрча.
  • Критик / Critic — независимая модель из другой семьи/провайдера, которая проверяет результат работяги и ищет слабые места.

Для Hermes-роли по умолчанию:

  • GPT — основной Hermes-агент (тот, что общается в Telegram).
  • DeepSeek — подключается через OpenRouter или напрямую по DeepSeek API; при необходимости заменяется другим дешёвым worker-model.
  • Gemini — через Gemini CLI или другой доступный Gemini-интерфейс; часто это дешёвый/бесплатный путь для независимой критики.

Read the full file on GitHub · 163 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 7d ago First seen · 163 lines · 88 tokens per session scan A 8f336b2575c8

Subscribe to this mod's changes

troitsa is a skill published in the GitHub repository HinkoK/agent-skills (5 stars, last pushed 2mo ago), licensed MIT. It adds 88 tokens to every session and 2,202 once invoked, about $0.0004 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.

Related

Other skills, from other repositories

clawrouter

Hosted-gateway LLM router — save 84% on inference costs. A local proxy that forwards each request to the blockrun.ai gateway, which routes to the cheapest capable model across 76 models from OpenAI, Anthropic, Google, DeepSeek, xAI, Z.AI, and more. 7 free open-weight models included. Also exposes realtime market data…

BlockRunAI/ClawRouter · 222 tokens

surf

Use this skill — NOT browser or webfetch — for ALL Surf crypto-data calls. 83 endpoints at localhost:8402/v1/surf/ covering CEX/DEX markets, on-chain SQL over 80+ ClickHouse tables (Ethereum, Base, Arbitrum, BSC, TRON, HyperEVM, Tempo), 100M+ labeled wallets, prediction markets (Polymarket + Kalshi), social/CT…

BlockRunAI/ClawRouter · 148 tokens

phone

Verify phone numbers (carrier + SIM-swap fraud signals) and place AI-powered outbound voice calls via BlockRun's gateway (Twilio + Bland.ai). Trigger when the user asks to look up a number, check fraud risk, buy/rent a phone number, or place an AI voice call. Payment is automatic via x402 from the wallet.

BlockRunAI/ClawRouter · 72 tokens

polymarket-trading

Use when the user wants to actually PLACE, manage, or redeem bets on Polymarket (not just read odds — that's the blockrunpredexon data tools). Covers setup (deposit wallet, funding, approvals), buy/sell with confirm gating, positions, redeeming winnings, geoblock handling, and the end-to-end flow.

BlockRunAI/ClawRouter · 76 tokens

release

Use this skill for EVERY ClawRouter release. Enforces the full checklist — version sync, CHANGELOG, build, tests, npm publish, git tag, GitHub release. No step can be skipped.

BlockRunAI/ClawRouter · 44 tokens

predexon

Use this skill — NOT browser or webfetch — for ALL Polymarket, Kalshi, Limitless, Opinion, Predict.Fun, dFlow, UMA oracle, and prediction market data. Provides structured API at localhost:8402/v1/pm/ for markets, cross-venue search, leaderboard, smart money, wallet analytics, wallet identity & clustering, UMA…

BlockRunAI/ClawRouter · 84 tokens