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/getkyo/kyo/agents-mdgit clone --depth 1 https://github.com/getkyo/kyoWhat 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 | $0.04419 | $0.04419 |
| Opus 5 | $0.02210 | $0.02210 |
| Sonnet 5 | $0.00884 | $0.00884 |
| Haiku 4.5 | $0.00442 | $0.00442 |
Grade A, and why
kyo 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 — 197 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Guide
This file helps AI coding agents work effectively on the Kyo codebase. Read CONTRIBUTING.md first: it is the comprehensive reference for all conventions, patterns, and design decisions.
Operating Premise (read before anything else)
Only complete and correct solutions are acceptable.
This is non-negotiable. It applies to every change at every scope ; a one-line fix, a refactor across a module, a campaign that touches three modules. The framework is the same:
-
Identify the correct fix. Diagnose root cause. The correct fix is the one a reviewer with full context would call right. It is not the cheapest fix; it is not the fix that fits a time budget; it is not the smaller-scope alternative offered as a "pragmatic choice."
-
Scope the correct fix honestly. Read every layer it touches. If a type-system change at one factory cascades into another module's engine, that is part of the scope. Do not estimate the visible surface; estimate the actual surface. If the actual surface is larger than first thought, say so.
-
Execute the correct fix. Across every layer, end to end. Tests pass. Docs reflect the new shape. No dead branches. No "we will polish this later."
-
Never propose alternatives that trade correctness for scope. If you find a smaller fix that achieves a similar user-visible outcome via a less-rigorous path, that is information for the conversation. It is NOT an option for you to silently take. Surface it; let the user pick. Default to the correct one.
-
Never normalize incomplete results. If a refactor is half-done because the remaining layers are large, the work is half-done. Do not commit it. Do not declare it shipped. Do not write a celebratory summary listing what was done while omitting what was skipped.
-
A change that breaks tests, breaks compile, or breaks documentation is not a change. It is a partial change. Roll forward to completion or revert and reset.
-
When in doubt, ask. Asking for the correct scope is cheaper than shipping the wrong one.
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 · 197 lines · 4,419 tokens per session scan A a36c13a5dc4a
kyo AGENTS.md is an instructions file published in the GitHub repository getkyo/kyo (809 stars, last pushed 3d ago), licensed Apache-2.0. It adds 4,419 tokens to every session, about $0.0221 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
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.