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 skills add rodrigorpassos/claude-codex-failover --skill failover-crossovergit clone --depth 1 https://github.com/rodrigorpassos/claude-codex-failoverWrote 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/skills/rodrigorpassos/claude-codex-failover/failover-crossover)<a href="https://agentmods.dev/skills/rodrigorpassos/claude-codex-failover/failover-crossover"><img src="https://agentmods.dev/badge/skills/rodrigorpassos/claude-codex-failover/failover-crossover.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 5.1 | $0.00278 | $0.01896 |
| Opus 5 | $0.00139 | $0.00948 |
| Sonnet 5 | $0.00056 | $0.00379 |
| Haiku 4.5 | $0.00028 | $0.00190 |
Grade A, and why
failover-crossover 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 7d 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 — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
failover-crossover — lado CLAUDE (Opus)
Skill crossover de failover manual, espelho inverso de codex/skills/failover-crossover
(lado Codex). Quem aciona é o lado UP: esta skill roda no Claude, quando o GPT/Codex caiu
e o Opus está vivo. O modelo de resiliência não precisa de processo externo porque a skill que executa
é a do lado que sobreviveu.
Fontes (ler de 1ª mão ao acionar): docs/contract.md (§4 failover, §5 isolamento) ·
docs/runbook.md (prompts P3/P4) · docs/adr/0001 (dois orquestradores) e docs/adr/0002 (sem
degradar modelo). Gate de reconciliação (contexto separado): para scope=harness-meta
(governança), um subagente revisor adversarial em contexto separado; para scope=project
(código), revisão em instância/sessão distinta apropriada ao código.
Convenção de instalação: ao instalar no seu harness, copie este
SKILL.mdpara~/.claude/skills/failover-crossover/e os docs citados para onde preferir (ex.:~/.claude/docs/). Os paths~/.claude(casa do Claude Code),~/.codex(casa do Codex CLI) e~/.agents/skills(skills do Codex) são as localizações padrão dos harnesses. As branchesfailover/harnesssão nomes de exemplo — ajuste aos da sua topologia.
Parâmetros canônicos (mudança de estado)
Invocação: failover-crossover <event> [scope=<project|harness-meta>] [context="<...>"]
event— OBRIGATÓRIO, canônico; a transição de estado observada. Vocabulário crossover compartilhado ={opus-down, opus-up, gpt-down, gpt-up}. Esta skill (Claude) trata apenas:gpt-down— GPT/Codex indisponível, Opus UP → Claude ASSUME a sessão do Codex e continua (runbook P3).gpt-up— GPT/Codex voltou → o Claude disponibiliza o estado (versionado/handoff) e o Codex realinha a própria casa em sessão nativa (runbook P4); o Claude não executa P4 (ponte read-only em~/.codex).- (
opus-down/opus-uppertencem ao lado Codex — recusar com ponteiro à skill espelho.)
scope— canônico; fixa a invariante §5 (destino do trabalho):project→ trabalho vai ao repo/branch do projeto.harness-meta→ meta-trabalho no harness → branchfailover(e o Claude primário reconcilia aharnessa partir dela).
context— texto livre; o que a sessão interrompida fazia / objetivo / onde parou (insumo da recuperação).
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.
- 7d ago First seen · 88 lines · 278 tokens per session scan A bb77d0a7840e
failover-crossover is a skill published in the GitHub repository rodrigorpassos/claude-codex-failover (1 stars, last pushed 2mo ago), licensed MIT. It adds 278 tokens to every session and 1,896 once invoked, about $0.0014 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 skills, from other repositories
frontend-state-management
SocialApp user frontend state management — TanStack Query, Zustand, IndexedDB persistence, cache invalidation, multi-account isolation.
dev-orchestrator
Full-stack development orchestrator -- always-on, 16 phases (0-15): read-docs -> understand -> architect -> pseudocode -> plan -> contract -> validate -> implement -> evaluate -> verify -> test -> goals -> review -> critic -> document -> report.
bot-reviewer
Review chat bot code (Telegram, Discord, Slack) for callback safety, state machines, rate limits, and production readiness.
silent-failure-hunter
Detects swallowed errors, empty catch blocks, log-and-forget patterns, and fallback masks that hide failures. Zero tolerance for silent failures. Severity-graded output with concrete fixes per language.
writing-hooks
How to write Claude Code hooks — lifecycle events, exit codes, profiles, jq/node parsing, settings.json wiring, and testing.
gan-evaluator
Runs Playwright against gan-generator's output, scores against the plan's rubric, returns concrete failures. Step 3 of 3 in GAN harness — the adversarial verification step.