Borrowing it
Nothing to install: this file belongs to Black-coffe/vulyk. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/Black-coffe/vulyk/main/.claude/commands/vulyk-models.mdgit clone --depth 1 https://github.com/Black-coffe/vulykWrote 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/commands/black-coffe/vulyk/vulyk-models)<a href="https://agentmods.dev/commands/black-coffe/vulyk/vulyk-models"><img src="https://agentmods.dev/badge/commands/black-coffe/vulyk/vulyk-models/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/commands/black-coffe/vulyk/vulyk-models"><img src="https://agentmods.dev/badge/commands/black-coffe/vulyk/vulyk-models.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00056 | $0.00815 |
| Opus 5.5 | $0.00022 | $0.00326 |
| Sonnet 5.5 | $0.00011 | $0.00163 |
| Haiku 4.5 | $0.00006 | $0.00081 |
Grade A, and why
vulyk-models 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 10d 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.
What it actually says
Let the owner choose which model family does what on THIS machine. One question, then one line. The choice lives in
one file per machine that only scripts/model-profile.sh --set writes; this command never picks for the owner.
- Read the state with plain Bash calls from the hive root:
bash scripts/model-profile.sh --recommendprints<profile> <plan>, andbash scripts/model-profile.sh --explainprints the active profile and who does what. If the script is missing or fails, say so in one line and stop. - Language: the constitution's
| Owner language |row (CLAUDE.vulyk.mdfirst, thenCLAUDE.md). While it still reads<fill in, use the language the owner writes in. Every word the owner sees below is in that language. - No
AskUserQuestionin your tool list (claude -p): write nothing and print one line, «run/vulyk-modelsin an interactive session», in the owner's language. Stop. - Ask ONE
AskUserQuestionwith the three profiles. Order: the profile--recommendnamed goes first, its label ends with «(Рекомендую)» (in the owner's language), and its description opens with the reason: the detected plan (<plan>;unknownmeans it could not be read, so «как сейчас» is the safe pick). The other two follow in the order below. Each option's description is one or two plain sentences; its preview says what each family does and what it costs in limits. Family names only: no percentages, no model versions.sonnet-builds, «как сейчас»: Sonnet writes the code of big tasks and the routine; Opus plans, judges and leads the session; Fable holds only the final gate on the riskiest calls, where the plan carries it; Haiku keeps the paperwork. Cost: the plan's usual mix; Fable spends a little of its own share of the Max weekly limit.opus-builds, «Opus пишет, Fable судит»: Opus writes the code of big tasks and leads the session; Haiku writes the routine stories; Fable plans, judges every review and holds the gate; Sonnet takes only the second Tier 4 review. Cost: Opus is the heaviest on code, so its limit runs out sooner; Fable spends from its share of the Max weekly limit, up to half of it. When--recommenddid not nameopus-builds(the plan does not carry Fable: Pro, standard seats, API, unknown), add: on this plan Fable is not inside the subscription and bills to usage credits.economy, «без Fable»: Sonnet writes the code of big tasks, Haiku writes the routine; Opus plans, judges, holds the gate and leads the session; Fable is not used. Cost: the lightest on limits and never touches Fable or credits.
- Write the answer:
bash scripts/model-profile.sh --set <name>. A free-text answer that names none of the three writes nothing; say so in one line. - Confirm in one line, in the owner's language: the profile now active, and that it can be changed any time with
/vulyk-modelsor in words («поменяй распределение моделей»).
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.
- 10d ago First seen · 35 lines · 56 tokens per session scan A 53a052e45d66
vulyk-models is a command published in the GitHub repository Black-coffe/vulyk (2 stars, last pushed today), licensed MIT. It adds 56 tokens to every session and 815 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-10-19.
Other commands, from other repositories
audit-agents-skills
Audit quality of agents, skills, and commands in a Claude Code project.
security-audit
Comprehensive security audit of your project AND Claude Code configuration. Analyzes secrets exposure, injection surfaces, dependencies, hook security, and produces a scored security posture assessment.
audit-quiz-coverage
Find quiz coverage gaps from recent guide/CHANGELOG/CC-releases changes and propose new questions.
worktree
Create a git worktree for isolated feature/fix work.
sync
Check guide/landing synchronization status.
init-docs
Fetch official Anthropic Claude Code docs and store as local baseline snapshot.