Borrowing it
Nothing to install: this file belongs to kagan-sh/kagan. 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/kagan-sh/kagan/main/AGENTS.mdgit clone --depth 1 https://github.com/kagan-sh/kaganWrote 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/kagan-sh/kagan/agents-md)<a href="https://agentmods.dev/instructions/kagan-sh/kagan/agents-md"><img src="https://agentmods.dev/badge/instructions/kagan-sh/kagan/agents-md/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/instructions/kagan-sh/kagan/agents-md"><img src="https://agentmods.dev/badge/instructions/kagan-sh/kagan/agents-md.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 | $0.01497 | $0.01497 |
| Opus 5.5 | $0.00599 | $0.00599 |
| Sonnet 5.5 | $0.00299 | $0.00299 |
| Haiku 4.5 | $0.00150 | $0.00150 |
Grade A, and why
kagan 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 11d 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 — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.
kagan: rules for every change
The spec on this branch wins over everything below: CONTEXT.md, docs/architecture.md, docs/plugin-api.md, docs/look.md.
Load these skills first
gpui-kit: its Coding Guides are normative for state ownership, entities, public API, naming and tests.gpui-kit-design-guides: for any UI.rust-patterns: for Rust in general.
Where a skill disagrees with the spec, the spec wins. Known overrides:
- No tokio. Async runs on GPUI's executors (
cx.spawn,cx.background_spawn); the ACP crate is runtime-agnostic. - Shared UI state lives in GPUI entities (
Entity<T>,cx.notify(), events), notArc<Mutex<T>>.
Laconic
- Write the smallest code that does the ticket. No speculative options, no trait with one implementor, no builder for three fields, no wrapper that only forwards.
- Delete what you replace. No shims, no dead code, no
#[allow(dead_code)]. - No comments except where the code would surprise a reader. No doc comments on private items.
- Names come from
CONTEXT.md. One word per concept.
Idiomatic Rust
- Errors:
Resultand?;thiserrorin library crates,anyhowonly in thekgbinary. Nounwrap/expectoutside tests unless the invariant is stated in the message. - States are enums, matched exhaustively. Parse at the boundary, pass typed values inside.
- Borrow before you clone; iterators before index loops;
impl Traitin argument position for single-use generics. - Visibility is the smallest that compiles:
pub(crate)by default,pubonly for the crate's interface indocs/architecture.md.
GPUI
- Views implement
Render; stateless pieces implementRenderOnce. No I/O and no heavy work inrender. - Retained state (inputs, lists, subscriptions, tasks) is owned by an entity and kept alive by a field; drop a
TaskorSubscriptiononly to cancel it. - Commands are GPUI actions with keybindings, and every user-facing one is also an
Op. ElementIds are stable across renders.- Colors and spacing come from the Workbench theme tokens (
docs/look.md), never literals in views. - UI behavior is tested with GPUI's test context through the view's interface.
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.
- 11d ago Changed · -43 lines · -554 tokens per session b006debeed9a
- 1mo ago First seen · 118 lines · 2,051 tokens per session scan A ec09ce253dde
kagan AGENTS.md is an instructions file published in the GitHub repository kagan-sh/kagan (106 stars, last pushed 11d ago), licensed MIT. It adds 1,497 tokens to every session, about $0.0060 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 2026-08-31.
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).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
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 copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deno copilot-instructions.md
Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.