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/astandrik/local-ydb-toolkit/agents-mdgit clone --depth 1 https://github.com/astandrik/local-ydb-toolkitWrote 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/astandrik/local-ydb-toolkit/agents-md)<a href="https://agentmods.dev/instructions/astandrik/local-ydb-toolkit/agents-md"><img src="https://agentmods.dev/badge/instructions/astandrik/local-ydb-toolkit/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 5.1 | $0.02200 | $0.02200 |
| Opus 5 | $0.01100 | $0.01100 |
| Sonnet 5 | $0.00440 | $0.00440 |
| Haiku 4.5 | $0.00220 | $0.00220 |
Grade A, and why
local-ydb-toolkit 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 6d 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 — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
local-ydb-toolkit - Agent Instructions
Build & Run
npm ci- Install exact workspace dependencies for CI-parity checks.npm install- Install workspace dependencies during local development.npm run build- Build@local-ydb-toolkit/coreand then@astandrik/local-ydb-mcp.npm run build -w @local-ydb-toolkit/core- Build only the core package.npm run build -w @astandrik/local-ydb-mcp- Build only the MCP server package.npm run typecheck- Type-check all workspaces with package-specifictsconfig.typecheck.jsonfiles.npm run compare:formats -w @astandrik/local-ydb-mcp- Compare JSON and TOON response sizes when changing MCP response formatting.node packages/mcp-server/dist/index.js- Start the stdio MCP server after a successful build.
Overview
This repository contains a reusable Codex skill for operating local-ydb deployments, a TypeScript npm workspace that exposes the same operations through a local stdio MCP server, and CI assets for the astandrik/setup-local-ydb GitHub Action. The runtime package is split between a core operation library and an MCP server wrapper.
Testing
npm test- Run all Vitest tests matchingpackages/**/*.test.ts.npm test -- packages/core/test/validation.test.ts- Run one core test file.npm test -- packages/mcp-server/test/tools.test.ts- Run one MCP server test file.npm test -- -t "config validation"- Run tests matching a Vitest name pattern.
Run npm test and npm run typecheck before handing off TypeScript changes.
GitHub CI uses Node 24 and runs npm ci, npm run build, npm test, and npm run typecheck.
The live MCP integration workflow also builds the server, starts YDB through astandrik/setup-local-ydb@v1, and runs node scripts/ci/verify-live-mcp-server.mjs; treat that as Docker/YDB integration coverage, not a default local check.
Project Structure
packages/core/- TypeScript library for config validation, command execution, local-ydb checks, and mutating operation plans.packages/mcp-server/- Stdio MCP server that registers publiclocal_ydb_*tools and delegates to@local-ydb-toolkit/core.packages/mcp-server/src/tools/- MCP tool registry, argument schemas, input schemas, instructions, and handler context.packages/mcp-server/src/prompts.ts- Guided MCP prompt templates for diagnosis, bootstrap, upgrades, auth hardening, and storage reduction.packages/mcp-server/src/response-format.tsandpackages/mcp-server/src/responses.ts- MCP response shaping, including optional TOON text content while keepingstructuredContentJSON.skills/local-ydb/- Codex skill content, references, agent metadata, scripts, and assets.examples/local-ydb.config.example.json- Example local and SSH target profiles; keep real secrets and host details out of committed configs.scripts/ci/- GitHub Actions verification scripts for live MCP and setup-local-ydb smoke coverage..github/workflows/- CI, live MCP integration, setup-local-ydb smoke, and release/publish workflows.server.json- Official MCP Registry metadata forio.github.astandrik/local-ydb-mcp.MCP_TOOL_TEST_SCENARIOS.md- Manual MCP tool scenarios.vitest.config.ts- Vitest configuration and workspace alias for@local-ydb-toolkit/core.../local-ydb-toolkit-ui- Sibling checkout for the public website when present; keep website-facing product metadata in sync when this repo changes public discovery or install surfaces.
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.
- 6d ago First seen · 91 lines · 2,200 tokens per session scan A 63b21341e25c
local-ydb-toolkit AGENTS.md is an instructions file published in the GitHub repository astandrik/local-ydb-toolkit (4 stars, last pushed yesterday), licensed MIT. It adds 2,200 tokens to every session, about $0.0110 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
appwrite AGENTS.md
AGENTS.md instructions for appwrite/appwrite, covering appwrite, commands, stack, layout and libraries.
clawhub AGENTS.md
AGENTS.md instructions for openclaw/clawhub, covering repository guidelines, project structure & module organization, durable intent & specs, build, test, and development commands and coding style & naming conventions.
netdata AGENTS.md
AGENTS.md instructions for netdata/netdata, covering agents.md, goals, requirement language, working with the user and development principles.
selfhost-ai CLAUDE.md
Claude Code instructions for kossakovsky/selfhost-ai, covering claude.md, project overview, core architecture, key files and installation flow.
she-love-me CLAUDE.md
Instructions for 863401402/she-love-me, covering claude.md and 唯一工作流.
homelab-monitor CLAUDE.md
Claude Code instructions for SikamikanikoBG/homelab-monitor, covering claude.md and conventions.