Borrowing it
Nothing to install: this file belongs to krzysztofdudek/RatatoskrSkill. 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/krzysztofdudek/RatatoskrSkill/main/CLAUDE.mdgit clone --depth 1 https://github.com/krzysztofdudek/RatatoskrSkillWrote 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/krzysztofdudek/ratatoskrskill/claude-md)<a href="https://agentmods.dev/instructions/krzysztofdudek/ratatoskrskill/claude-md"><img src="https://agentmods.dev/badge/instructions/krzysztofdudek/ratatoskrskill/claude-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/krzysztofdudek/ratatoskrskill/claude-md"><img src="https://agentmods.dev/badge/instructions/krzysztofdudek/ratatoskrskill/claude-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 | $0.01352 | $0.01352 |
| Opus 5 | $0.00676 | $0.00676 |
| Sonnet 5 | $0.00270 | $0.00270 |
| Haiku 4.5 | $0.00135 | $0.00135 |
Grade A, and why
RatatoskrSkill CLAUDE.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 today.
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 — 31 lines — stays where its author put it; the contents beside it link to each section on GitHub.
RatatoskrSkill
Purpose
This repository exists solely so the author can develop and version the ratatoskr skill. The canonical file is skills/ratatoskr/SKILL.md — people install it as a Claude Code plugin (or copy that one file into their agent's skill dir). Nothing in this repo (CLAUDE.md, CHANGELOG.md, README.md, CI, etc.) may affect the skill's mechanics. All behavior must be self-contained in skills/ratatoskr/SKILL.md.
Ratatoskr is an add-on in the Yggdrasil family: it attaches to the agent, not to the graph, and works alone. It is the communication layer — the courier that keeps the agent talking to the user in plain words, never in code. It governs how the agent communicates on every turn, not how the work is sequenced. The family's core is Yggdrasil (the law: the architecture graph and the rails), Grain (surveys the terrain: the first graph, mined from a repository's own code and history) and Horde (the software house on the law: zero standing roles, a worker per ticket and a one-shot architect); adoption runs Grain first, day zero, then Yggdrasil as the long-term core, then Horde once a mission needs more hands than one agent. Ratatoskr, Urd and Researcher are the three add-ons beside them, each with no dependency of its own.
Plugin scaffolding
This repo is installable as a Claude Code plugin and as a GitHub Copilot CLI plugin. Layout mirrors the convention used by sibling repos (e.g. UrdSkill, ResearcherSkill):
.claude-plugin/plugin.json— plugin manifest (name, version, description, keywords).versionhere MUST match the latest released version inCHANGELOG.mdand is bumped together with it..claude-plugin/marketplace.json— single-plugin marketplace listing for Claude Code, so the repo can be added via/plugin install ratatoskr@ratatoskr-marketplace..github/plugin/marketplace.json— single-plugin marketplace listing for GitHub Copilot CLI (Copilot reads this path), so the repo can be added viacopilot plugin marketplace add krzysztofdudek/RatatoskrSkillthencopilot plugin install ratatoskr@ratatoskr-marketplace. Mirrors the Claude listing but additionally carriesversion, askillsarray (./skills/ratatoskr), and ahooksfield pointing at./hooks/copilot-hooks.json(the Copilot-format session-start hook). Its pluginversionMUST be kept in lockstep withplugin.json..codex-plugin/plugin.json— plugin manifest for OpenAI Codex CLI (Codex reads the plugin manifest only from.codex-plugin/). Bundles the skill via"skills": "./skills/"; Codex discovers the marketplace from the existing.claude-plugin/marketplace.json(its legacy-compatible path), so the repo installs viacodex plugin marketplace add krzysztofdudek/RatatoskrSkillthencodex plugin install ratatoskr@ratatoskr-marketplace. ItsversionMUST be kept in lockstep withplugin.json. Codex needs no hook wiring of its own: it auto-discovers the defaulthooks/hooks.json, reuses Claude'sSessionStart/hookSpecificOutputschema, and setsCLAUDE_PLUGIN_ROOTfor compatibility, so the bundled Claude hook fires under Codex unchanged..cursor-plugin/plugin.json— plugin manifest for Cursor (single-plugin-at-root: manifest at the repo root, no Cursor marketplace file; components are auto-discovered, soskills/ratatoskr/is picked up automatically). Installed locally via~/.cursor/plugins/local/or published to the Cursor Marketplace. Carries ahooksfield pointing athooks/cursor-hooks.json(the Cursor-format session-start hook). ItsversionMUST be kept in lockstep withplugin.json.hooks/— session-start activation hooks, one per platform we integrate with. Activation-only: each injects a short instruction so the agent applies the skill from the first turn and re-asserts it after compaction; they encode no rules of their own, so the self-containment rule holds — all behavior still lives inSKILL.md. One shared script,hooks/ratatoskr-session-start.sh, emits the right stdout envelope based on its$1argument (claude/codex→hookSpecificOutput.additionalContext;copilot→ flatadditionalContext;cursor→ snake_caseadditional_context). The per-platform configs:hooks/hooks.json(ClaudeSessionStart, also auto-discovered by Codex),hooks/copilot-hooks.json(CopilotsessionStart, wired via the.github/plugin/marketplace.jsonhooksfield, uses${PLUGIN_ROOT}), andhooks/cursor-hooks.json(CursorsessionStart, wired via the.cursor-plugin/plugin.jsonhooksfield). The message lives once, in the script.skills/ratatoskr/SKILL.md— the canonical skill body. Editing this file IS editing the skill.
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.
- today Changed · +55 tokens per session 82decd584d52
- 3d ago Changed · +54 tokens per session aaafedcd816c
- 12d ago First seen · 31 lines · 1,243 tokens per session scan A 5d990bfb3c93
RatatoskrSkill CLAUDE.md is an instructions file published in the GitHub repository krzysztofdudek/RatatoskrSkill (2 stars, last pushed today), licensed MIT. It adds 1,352 tokens to every session, about $0.0068 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
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.
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.
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.
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).
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).