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/anzellai/sky/agents-mdgit clone --depth 1 https://github.com/anzellai/skyWhat 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 | $0.12356 | $0.12356 |
| Opus 5 | $0.06178 | $0.06178 |
| Sonnet 5 | $0.02471 | $0.02471 |
| Haiku 4.5 | $0.01236 | $0.01236 |
Grade C, and why
sky AGENTS.md scanned grade C with 1 finding 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
builds clean from a wiped slate (`rm -rf sky-out .skycache .skydeps && sky How it starts
The opening of the file, as written. The whole thing — 710 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Agent-agnostic guide for AI tools (Claude, Copilot, Cursor, Codex, …) working in the Sky repository. It is the source of truth; tool-specific files layer on top of it (Claude Code reads
CLAUDE.md, which imports this file and adds Claude-Code-only operational rules).Progressive disclosure. This file is deliberately lean. It teaches the language, the app-building decisions, and the non-negotiable rules, then points you at two authoritative, always-current sources for depth:
sky doc <Module>— the live stdlib API (typed signatures + summaries + examples), generated from the stdlib source, so it never drifts. Prefer it over any hand-copied API table.sky doc --listenumerates every module;sky doc --serveopens a browsable server.docs/— deep dives (architecture, Sky.Live, Std.Db, Std.Ui, auth, tooling). The map is in the Deep dives table below.
What Sky is
Sky is an Elm-family, purely-functional language that compiles to typed Go. One language for the whole stack: web UIs (Sky.Live), HTTP/JSON APIs, CLIs, TUIs, desktop apps, background jobs. The design goal is "if it compiles, it works" — no user-written FFI, no nulls, no runtime panics from well-typed code, clear errors, batteries-included stdlib.
The compiler is the Rust rewrite (cargo workspace at rust/). The retired
Haskell compiler lives under legacy-haskell-compiler/ and serves as a
differential oracle (sky-out/sky) the Rust output is checked against
byte-for-byte. Current line: v0.23.x.
Language essentials
Sky's surface is Elm. If your training on Elm is thin, read this section carefully — it is the part that catches models out.
module Main exposing (main)
import Sky.Core.Prelude exposing (..) -- Result/Maybe/identity/… autoloaded
import Sky.Core.List as List
import Std.Log exposing (println)
type alias User = { name : String, age : Int } -- record alias
type Msg = Increment | Decrement -- tagged union (ADT)
greet : User -> String -- type annotation (optional but preferred)
greet u =
"Hi " ++ u.name
update : Msg -> Int -> Int
update msg count =
case msg of -- case is exhaustiveness-checked
Increment -> count + 1
Decrement -> count - 1
main =
println (greet { name = "Ada", age = 40 })
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 · 710 lines · 12,356 tokens per session scan C ba639f322c8b
sky AGENTS.md is an instructions file published in the GitHub repository anzellai/sky (432 stars, last pushed 3d ago), licensed Apache-2.0. It adds 12,356 tokens to every session, about $0.0618 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
compiler-explorer AGENTS.md
Instructions for compiler-explorer/compiler-explorer, covering agents.md, build & test commands, important workflow requirements, style guidelines and architecture guidelines.
compiler-explorer copilot-instructions.md
Instructions for compiler-explorer/compiler-explorer: For each answer to the user, evaluate your level of confidence in the correctness of the answer on a scale from 1 to 10, where 1 is very uncertain and 10 is absolutely certain. If your confidence level is below 8, state so and suggest ways to verify the answer.
neva AGENTS.md
AGENTS.md instructions for nevalang/neva, covering neva engineering guide, project model, repository map, documentation and ai engineering harness.
agent-lsp CLAUDE.md
Instructions for blackwell-systems/agent-lsp, a project described as: MCP server that orchestrates language servers into agent-native workflows. 65 tools, 30 CI-verified languages.
GoSQLX CLAUDE.md
Instructions for ajitpratap0/GoSQLX, covering claude.md, project overview, architecture, core components and token processing pipeline.
dingo CLAUDE.md
Claude Code instructions for MadAppGang/dingo, covering claude ai agent instructions - dingo project, 🚨🚨🚨 stop: read this before any implementation 🚨🚨🚨, the architectural principle (understand this first), why this rule exists and ✅ required approaches (the only correct ways).