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 skills add serejaris/personal-corp-os --skill fable-ruki-agentygit clone --depth 1 https://github.com/serejaris/personal-corp-osWrote 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/skills/serejaris/personal-corp-os/fable-ruki-agenty)<a href="https://agentmods.dev/skills/serejaris/personal-corp-os/fable-ruki-agenty"><img src="https://agentmods.dev/badge/skills/serejaris/personal-corp-os/fable-ruki-agenty/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.
<a href="https://agentmods.dev/skills/serejaris/personal-corp-os/fable-ruki-agenty"><img src="https://agentmods.dev/badge/skills/serejaris/personal-corp-os/fable-ruki-agenty.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 4 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Agent Snooping · line 32 Skill reads from agent configuration directories (.claude/, .codex/, .gemini/). These directories may contain API keys, personal settings, and other credentials that the skill has no legitimate need to access.Fix: Remove all code or instructions that access agent configuration directories (.claude/, .codex/, .gemini/). If configuration values are needed, pass them explicitly as parameters or environment variabl
- high Excessive Agency · line 60 Skill selects an external model or provider that may use a different account or billing plan than the operator expects. Undisclosed model switches can cause unexpected cost or quota consumption.Fix: Remove the model/provider override or disclose it prominently and require explicit operator approval before invoking an external coding CLI or billed model.
- high Excessive Agency · line 60 Skill selects an external model or provider that may use a different account or billing plan than the operator expects. Undisclosed model switches can cause unexpected cost or quota consumption.Fix: Remove the model/provider override or disclose it prominently and require explicit operator approval before invoking an external coding CLI or billed model.
- medium analysis-evasion · line 1 Suspicious Unicode normalization or mixed-script contentFix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
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 5.1 | $0.00116 | $0.07176 |
| Opus 5 | $0.00058 | $0.03588 |
| Sonnet 5 | $0.00023 | $0.01435 |
| Haiku 4.5 | $0.00012 | $0.00718 |
Grade B, and why
fable-ruki-agenty scanned grade B with 1 finding 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 12d 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
Кодинг, код-ревью, gh-операции, CLI-механика, DoD-верификация и укладка файлов конвейер отдаёт **Codex** (плагин `codex`, runtime codex-companion). Модель — **`--model gpt-5.5`, всегда явно** (не полагаться на дефолт `~/ How it starts
The opening of the file, as written. The whole thing — 291 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Фейбл руки-агенты
Зачем этот режим
Токены Fable — самый дорогой ресурс сессии (лимиты урезаны). Поэтому Fable делает только то, что не может сделать рука дешевле: думает, резолвит развилки и пишет спеки. Всё остальное делают руки: кодинг, код-ревью, gh-операции, CLI-механика, DoD-верификация и укладка файлов — Codex (gpt-5.5, плагин codex, § Codex-руки); разведка, чтение и скауты — Grok-воркеры через Orca (§ Orca-руки). Каждый tool call, который могла сделать рука, — сожжённый лимит.
Sonnet и Haiku не использовать ни для каких ролей (решение founder 04.07.2026).
Граница проходит по суждению, а не по стоимости токена: субагентам уходят глаза и руки, но никогда — голова. Выбор из вариантов, приоритеты, синтез рисёча в выводы, редакционные/продуктовые/архитектурные решения, тексты стратегий и спек — Fable делает сам. Суждение — то, ради чего Fable в сессии; решение, делегированное дешёвой модели, — не экономия, а подмена самого дорогого самым дешёвым.
Жёсткие запреты
- Не писать код и не править файлы репо. Не вызывать Bash/gh вообще —
gh issue edit/view, CLI-механика и коммиты уходят в Codex/форвардер. Fable оставляет себе только: мышление, спеки, резолв развилок, диспатч агентов. Codex-задача диспатчится через Agent tool (codex:codex-rescue); Bash у Fable не появляется нигде. - Не читать кодовую базу самому. Нужен контекст — Grok-скаут через Orca с конкретным вопросом и форматом ответа, назад приходит summary.
- Не диспатчить задачу без готовой спеки в теле issue. «Сделай #2» по сырому issue — главный анти-паттерн: исполнитель начнёт разведку сам и сожжёт токены повторно.
- Никакого ultrathink / xhigh по умолчанию. High достаточно; xhigh — только для резолва действительно сложной архитектурной развилки. High руке — только если верификатор дважды завернул задачу и диагноз «недодумала», не по умолчанию.
- Не отдавать суждение. Скаут — разведчик: приносит факты (file:line, URL, числа, цитаты, полные списки с атрибутами), решает Fable. В промпте скауту запрещены «выбери», «реши», «предложи», «оцени, как лучше», «сделай вывод»; разрешены «найди», «перечисли», «измерь», «процитируй», «сверь», «прогони». Нужен выбор из N объектов — скаут возвращает все N с объективными атрибутами (даты, размеры, метрики), отбор делает Fable. Суждение не достаётся никому, кроме Fable, — ни при каком размере задачи.
What ships with it
3 files 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.
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.
- 12d ago First seen · 291 lines · 116 tokens per session scan B 6093b87c3c33
fable-ruki-agenty is a skill published in the GitHub repository serejaris/personal-corp-os (225 stars, last pushed 16d ago), licensed MIT. It adds 116 tokens to every session and 7,176 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
milestone-runner
Use when asked to "run milestones", "execute EXECUTIONPROMPTS.md", "continue the milestone sequence", "run independent milestones in parallel", "merge the stack", or "reconcile M5" after prior work changes the current design. Not for generating plan files; use plan-prompts. Not for repairing one stack; use stacked-prs.
task-decomposer
Produces phased task boards from feature requests: dependency-mapped work items, parallelization flags, risk flags, edge cases, test matrices. Triggers on: "decompose this feature", "task breakdown with dependencies", "phased implementation plan", "work breakdown structure". NOT for discovering which decisions are…
project-context-setup
Scaffolds per-repository agent context so coding agents share the same issue tracker rules, triage label vocabulary, domain glossary, ADR layout, and handoff conventions. Triggers on: "set up project context", "configure agent docs", "create CONTEXT.md", "setup agent workflow", "agent issue tracker setup", "triage…
github
GitHub CLI operations via gh for issues, PRs, Actions, releases, and REST/GraphQL API with --json/--jq parsing. Triggers on: "create an issue", "submit a PR", "check CI status", "why did CI fail", "merge a PR", or pasted GitHub URLs.
slack
Slack workspace integration: 9 on-demand scripts (post, read, delete, search, stream, react, upload, channels, users) + dual-token support + RTS API search + chat streaming + AI Block Kit + Session Bridge + Claudicle unified launcher (Claude Agent SDK, soul engine, three-tier memory).
mx-flow
Full development workflow orchestrator: brainstorm → plan → worktree → convergent TDD/review loop → verify → PR. One human gate (spec approval); all other gates auto-proceed. Use when starting a feature or significant change from scratch. Usage: /mx-flow ; /mx-flow status [name]; /mx-flow finish .