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/kpavlov/tachyon/agents-mdgit clone --depth 1 https://github.com/kpavlov/tachyonWrote 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/kpavlov/tachyon/agents-md)<a href="https://agentmods.dev/instructions/kpavlov/tachyon/agents-md"><img src="https://agentmods.dev/badge/instructions/kpavlov/tachyon/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 | $0.00799 | $0.00799 |
| Opus 5 | $0.00400 | $0.00400 |
| Sonnet 5 | $0.00160 | $0.00160 |
| Haiku 4.5 | $0.00080 | $0.00080 |
Grade A, and why
tachyon 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 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.
How it starts
The opening of the file, as written. The whole thing — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Tachyon MCP — Agent Guide
Style
- Me Caveman. Talk short. Use emoji.
- Verdict first, then evidence. No preamble.
- Claims about code carry
file:lineproof. Verify in code, never answer from memory. - Status lists for checklists: ✅ done / 🔴 open, one row per item, evidence column.
- Emoji as markers: 🎯 goals, 🔴 breaking, 🐛 bugs, 🪶 polish/decisions, 🏹 order of battle, ⚠️ caveats, 🗑️ delete.
- Short sentences. No filler words. Dead docs go fire 🔥.
Project
Java 21+/Kotlin 2.2 MCP server. Java first, Kotlin adapts.
Fast Commands
Use IDE MCP for building/running tests if available. Otherwise see
CONTRIBUTING.md for make targets (build, test, lint,
format, ci, all).
- Use
mvn -qwhen only pass/fail matters
Parts
tachyon-bom— Maven BOM: pins consumer-facing module versions viadependencyManagementtachyon-core— Core: Netty HTTP/SSE, JSON-RPC, event log, MCP registriese2e— E2E tests viaio.modelcontextprotocol.sdk:mcp-corev2.0+conformance— Conformance via@modelcontextprotocol/conformance
Rules
- No cheating!
- Unattended
git commit/pushis prohibited. Always confirm with user. - TDD + SOLID. TachyonServer is SUT in unit tests.
- Tests: JUnit 6 + Kotest (Kotlin) / AssertJ (Java) + Awaitility.
@TempDirfor unit, port 0 for E2E. Prefer E2E, esp. long scenarios; unit only when E2E can't cover, drop unit if E2E already does. No tautologies. Many asserts per test. - Nullability: JSpecify
@Nullable/@NonNull.@NullMarkedat package level. - No comments in code unless needs to explain.
- Javadocs for public API required - this OSS library for users
git mvfor files.- Use MCP tools.
- Format: Check on
make lint, fix withmake format. - API/Registry design: see
docs/architecture/guidance.mdbefore adding/changing a handler SAM (sync/async shape, checked exceptions, descriptor bundling,_meta) or naming registry APIs (ServerBuildernouns,register/registerAsync/unregister,find,descriptors()). - Kotlin DSL (
tachyon-kotlin):- Each scope class gets its own
*Scope.ktfile. TachyonServerBuilderwrapsServerBuilderas the DSL receiver. Scope methods onTachyonServerBuilderuse clean names (info,capabilities,network,session) with zero Java member conflicts.- Method naming:
@DslMarkerextensions onTachyonServerBuilderfollow Java builder convention — keep names short and idiomatic (info { },capabilities { }, etc.). - Entry points:
TachyonServer(port) { }(builds + starts transport),buildServer { }(builds only, no transport). - Run Kotlin tests:
mvn test -pl tachyon-kotlin -am.
- Each scope class gets its own
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.
- 3d ago First seen · 49 lines · 799 tokens per session scan A 4da4c4ada9fc
tachyon AGENTS.md is an instructions file published in the GitHub repository kpavlov/tachyon (30 stars, last pushed 3d ago), licensed Apache-2.0. It adds 799 tokens to every session, about $0.0040 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.
Other instructions, from other repositories
mail-time AGENTS.md
Instructions for veliovgroup/mail-time, covering agents.md, mission, topology / tuning (read readme for use-cases), multiple instances — encouraged and one mail host: 2–8 servers.
mail-time CLAUDE.md
Instructions for veliovgroup/mail-time, covering claude.md, what this repo is, where to look first, mental model for changes and common pitfalls to avoid.
ndif CLAUDE.md
Instructions for ndif-team/ndif, covering claude.md — ndif agent guide, what ndif is, top-level layout, architecture at a glance and security sandbox (the highest-stakes area).
chimp AGENTS.md
Instructions for softwaremill/chimp, covering rules, 1 - protocol conformance, 2 - tests organization, 3 - conformance tests and 4 - documentation.
GameServerManager AGENTS.md
AGENTS.md instructions for GSManagerXZ/GameServerManager, a project described as: GameServerManager(简称GSManager)现代化游戏一键部署面板,专为简化游戏服务器的部署、管理和维护而设计;注重steam、MC游戏一键开服.
silkweave CLAUDE.md
Instructions for silkweave/silkweave, covering claude.md, commands, mcp inspector (connects to mcp stdio example via .mcp.json), architecture and packages.