tools

A Vovk.ts guide for creating tools that language models can call. It covers tools made from existing procedures as well as standalone tools such as calculators, file operations, or SDK wrappers.

In plain words
What is it for?
Use it to derive tools from procedures, define independent tools, add names and descriptions, define input and output schemas, and prepare tools for OpenAI, Anthropic, the Vercel AI SDK, or MCP servers.
Why use it?
It removes the need to manually reshape each backend operation into a model-callable tool and explains how different tool formats can work together.

Skill for Claude CodeCodex

Part of the vovk plugin — 15 skills shipped together

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.

agentmods
npx agentmods add skills/finom/vovk/tools
Any agent
npx skills add finom/vovk --skill tools
Clone the repo
git clone --depth 1 https://github.com/finom/vovk

Made for: Claude Code, Codex.

Or install vovk, the plugin that ships this one along with the rest of its 15 skills.

Per session 261 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 5,890 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00261 $0.05890
Opus 5 $0.00130 $0.02945
Sonnet 5 $0.00052 $0.01178
Haiku 4.5 $0.00026 $0.00589

Measured 3d ago against content hash 45d4e4c60634, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

tools 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.

skills/tools/SKILL.md · 490 lines

How it starts

The opening of the file, as written. The whole thing — 490 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Vovk.ts LLM tools

Vovk produces LLM-consumable tool definitions — { name, description, parameters, execute, … } — via two entry points:

  • deriveTools({ modules }) — turn existing procedures (controllers, RPC modules, OpenAPI mixins) into tools auto.
  • createTool({ name, description, inputSchema?, outputSchema?, execute, … }) — hand-build standalone tool, no procedure backing. Use when body is plain code (SDK wrappers, calculators, file ops).

Both shapes interoperable — mix in one array, feed to OpenAI / Anthropic / Vercel AI SDK / MCP servers.

Out of scope: procedure authoring (procedure), @operation for OpenAPI docs (openapi), third-party OpenAPI mixin setup (mixins). MCP server transport / hosting outside Vovk; for Next.js, recommended runtime is mcp-handler npm package — example below.

Import path note. Code samples import from 'vovk-client'composed client + js template default, re-exported from node_modules/.vovk-client. With ts template, import from composedClient.outDir (e.g. @/client). With segmented client, import from @/client/<segment>. Call shape identical. See rpc skill.

deriveTools — core shape

import { deriveTools } from 'vovk';
import { TaskRPC, PetstoreAPI } from 'vovk-client';
import UserController from '@/modules/user/user-controller';

const { tools, toolsByName } = deriveTools({
  modules: { UserController, TaskRPC, PetstoreAPI },
});

modules accepts record of:

  • Controllers — execute via .fn() in-process, no HTTP.
  • RPC modules (from vovk-client or @/client[/segment] per layout) — execute via HTTP using standard fetcher.
  • OpenAPI mixins (same client surface) — execute via HTTP against third-party API. See mixins skill.

Each module yields one tool per procedure with @operation schema, not hidden.

Return shape

const { tools, toolsByName } = deriveTools({ modules: { TaskRPC } });

// Array — feed to LLM SDKs
llm.chat({ tools: tools.map(t => ({ name: t.name, description: t.description, input_schema: t.parameters })) });

// Map — dispatch tool calls without scanning
const result = await toolsByName[call.name].execute(call.arguments);

Read the full file on GitHub · 490 lines

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. 3d ago First seen · 490 lines · 261 tokens per session scan A 45d4e4c60634

Subscribe to this mod's changes

tools is a skill published in the GitHub repository finom/vovk (52 stars, last pushed 6d ago), licensed MIT. It adds 261 tokens to every session and 5,890 once invoked, about $0.0013 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.

Related

Other skills, from other repositories

ocli-api

Turn any OpenAPI/Swagger API into CLI commands and call them. Search endpoints with BM25, check parameters, execute — no MCP server needed.

EvilFreelancer/openapi-to-cli · 34 tokens

react-performance

Optimize React/Next.js runtime performance — eliminate waterfalls, shrink bundles, cut unnecessary re-renders, and speed up rendering. Use when writing, reviewing, or refactoring components for speed, or chasing slow renders, large bundles, janky interactions, or high TTFB/LCP/INP.

clubpay/ronykit · 64 tokens

ronykit-framework

Orchestrates RonyKit service development using the ronyup MCP server (knowledge resources, prompts, scaffold tools). Use when the user mentions RonyKit, ronyup, EdgeServer, contracts, scaffolding a workspace or feature, implementing API handlers and services, frontend bootstrap, integration tests, or design documents…

clubpay/ronykit · 75 tokens

storybook

Author and maintain Storybook stories with CSF 3.0 best practices — args, decorators, parameters, and config. Use when creating or editing .stories. files, configuring .storybook/, or ensuring every UI component ships with stories.

clubpay/ronykit · 50 tokens

webmcp

Build agent-friendly web apps with the W3C Web Model Context API (WebMCP). Use when exposing frontend features as structured tools for in-browser AI agents, adding document.modelContext tools to React/Next.js apps, designing tool schemas for dashboards or admin UIs, or making existing UI automatable without DOM…

clubpay/ronykit · 69 tokens

composition-patterns

Design reusable React component APIs that scale — compound components, lifted state, generic context, explicit variants. Use when a component is growing boolean props (isThread, isEditing…), when building a component library, or when reviewing component architecture. Includes React 19 API changes.

clubpay/ronykit · 58 tokens