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/x0x0b/codex-launcher/agents-mdgit clone --depth 1 https://github.com/x0x0b/codex-launcherWrote 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/x0x0b/codex-launcher/agents-md)<a href="https://agentmods.dev/instructions/x0x0b/codex-launcher/agents-md"><img src="https://agentmods.dev/badge/instructions/x0x0b/codex-launcher/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 | $0.00619 | $0.00619 |
| Opus 5 | $0.00309 | $0.00309 |
| Sonnet 5 | $0.00124 | $0.00124 |
| Haiku 4.5 | $0.00062 | $0.00062 |
Grade A, and why
codex-launcher 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 4d 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 — 35 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository Guidelines
Project Structure & Module Organization
src/main/kotlin/com/github/x0x0b/codexlauncher/— plugin actions (*Action), terminal service, CLI arg builder, settings UI/options.src/main/resources/META-INF/plugin.xml— plugin metadata; icons live insrc/main/resources/icons/.src/test/kotlin/com/github/x0x0b/codexlauncher/— JUnit tests mirroring main packages (e.g.,CodexArgsBuilderTest,InsertPayloadResolverTest).- Build outputs land in
build/; packaged zips inbuild/distributions/.
Build, Test, and Development Commands
./gradlew buildPlugin— compile and package the IntelliJ plugin zip intobuild/distributions/../gradlew runIde— launch the sandbox IDE with the plugin for manual checks../gradlew test— run the JUnit suite; use--tests com.github.x0x0b.codexlauncher.*to target classes../gradlew clean— remove build outputs before a fresh run (optional).
Coding Style & Naming Conventions
- Kotlin 2.2, JVM 21; follow standard Kotlin style (4-space indent, trailing commas where helpful, immutable vals preferred).
- Public APIs and services should include concise KDoc; logger via
com.intellij.openapi.diagnostic.logger. - Classes/objects use PascalCase; enums like
Model,Mode; actions end withAction; tests end withTest. - Keep IntelliJ platform threading rules in mind (UI changes on EDT; services stay light).
Testing Guidelines
- Tests live under
src/test/kotlinand use JUnit 4 (@Test); mirror package paths. - Favor fast, isolated unit tests for CLI arg building, action routing, and terminal handling.
- When adding behavior, pair with a regression test; name methods descriptively (
methodName_condition_expected). - For manual verification, prefer
./gradlew runIdeto exercise toolbar actions and settings UI.
Commit & Pull Request Guidelines
- Follow recent Conventional Commit style seen in history (
feat:,fix:, optional issue refs like(#95)); keep subject in imperative, ≤72 chars. - Include PR summary, rationale, and testing notes; attach screenshots/GIFs for UI changes and mention affected IDE versions (targets 2024.2+).
- Update
plugin.xmlchange notes when user-facing behavior shifts; align version inbuild.gradle.ktsas needed.
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.
- 4d ago First seen · 35 lines · 619 tokens per session scan A 69935e06ff44
codex-launcher AGENTS.md is an instructions file published in the GitHub repository x0x0b/codex-launcher (53 stars, last pushed 5d ago), licensed MIT. It adds 619 tokens to every session, about $0.0031 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
android-tools-mcp CLAUDE.md
Instructions for amsavarthan/android-tools-mcp, covering android-tools-mcp, build, run / test, connect with claude code and architecture.
swttch CLAUDE.md
Claude Code instructions for Swttch/swttch, covering claude.md, 프로젝트 개요, ★ 핵심 원칙 — cli 동등성 & 공식 지원 비의존 (최우선, 위반 금지), 기록된 예외 and 아키텍처.
swttch AGENTS.md
AGENTS.md instructions for Swttch/swttch, covering agents.md, "agent" 용어 정의, 프로젝트 개요, 아키텍처 and 구현 기준.
skillOps-intellij AGENTS.md
AGENTS.md instructions for spmisha134/skillOps-intellij, covering skillops - agent guide, prerequisites, architecture overview, module structure and package structure rules.
skillOps-intellij CLAUDE.md
Claude Code instructions for spmisha134/skillOps-intellij, a project described as: IntelliJ plugin for creating and validating Codex, Claude and Gemini skills inside a project.
apm python.instructions.md
Python development guidelines.