Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/686f6c61/alfred-devnpx agentmods add commands/686f6c61/alfred-dev/luciusWrote 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/commands/686f6c61/alfred-dev/lucius)<a href="https://agentmods.dev/commands/686f6c61/alfred-dev/lucius"><img src="https://agentmods.dev/badge/commands/686f6c61/alfred-dev/lucius/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/commands/686f6c61/alfred-dev/lucius"><img src="https://agentmods.dev/badge/commands/686f6c61/alfred-dev/lucius.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 | $0.00017 | $0.01121 |
| Opus 5.5 | $0.00007 | $0.00448 |
| Sonnet 5.5 | $0.00003 | $0.00224 |
| Haiku 4.5 | $0.00002 | $0.00112 |
Grade A, and why
lucius 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 1mo 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 — 104 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/alfred-dev:lucius
Eres Alfred, orquestador del equipo. El usuario quiere una segunda opinión técnica
externa sobre el código de su proyecto. Activa a Lucius usando la herramienta
Agent con subagent_type: "alfred-dev:lucius".
Argumentos del usuario: $ARGUMENTS
Protocolo helper-first y modo headless
Antes de lanzar a Lucius, intentar Codex CLI o leer el repo, consume el prefetch determinista si existe:
python3 .claude/alfred-continuity.py consume-prefetch "$PWD" --expected lucius
Si el prefetch existe y devuelve salida, responde con esa salida y termina. Si no existe, prepara la revisión sin ejecutarla:
python3 .claude/alfred-continuity.py lucius "$PWD" --raw "$ARGUMENTS"
En modo headless (claude -p), SDK sin callback usable, auditoría automática o
si una herramienta indica que hay prefetch consumido, NO lances Agent, NO
ejecutes codex exec y NO presentes una revisión como hecha. Devuelve
LUCIUS_HEADLESS_START con directorio, scope, prerequisitos y siguiente paso.
Si el scope es inválido, devuelve LUCIUS_INVALID_SCOPE y termina sin lanzar
Codex CLI.
En sesión interactiva normal, puedes continuar desde esa preparación y entonces activar a Lucius con confirmación y preflight de Codex CLI.
Uso
/alfred-dev:lucius → audita el directorio actual
/alfred-dev:lucius src/ → audita un subdirectorio concreto
/alfred-dev:lucius --scope security → solo problemas de seguridad
/alfred-dev:lucius src/ --scope tests → tests en un subdirectorio
Scopes disponibles
| Scope | Qué analiza |
|---|---|
all (por defecto) |
Auditoría completa: seguridad, arquitectura, tests, rendimiento |
security |
OWASP Top 10, secretos, validación de entrada, CVEs |
tests |
Cobertura, casos borde, rutas de error sin test |
architecture |
Acoplamiento, responsabilidad única, dependencias circulares |
performance |
N+1, operaciones bloqueantes, cuellos de botella |
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.
- 1mo ago First seen · 104 lines · 17 tokens per session scan A 96481b64c4e7
lucius is a command published in the GitHub repository 686f6c61/alfred-dev (117 stars, last pushed 2mo ago), licensed MIT. It adds 17 tokens to every session and 1,121 once invoked, about $0.0001 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 2026-08-30.
Other commands, from other repositories
review-pr
Review a PR (GitHub or Bitbucket) — single reviewer by default, or a 3-panelist council with --council.
code-review
Run a comprehensive automated code review on staged changes or a specific branch/PR. Usage: /code-review [branch-name] or /code-review --staged.
init
Install the formatters this repository needs, with every command visible before it runs.
simplify
The over-engineering review: five tags (delete, stdlib, native, yagni, shrink), a mandatory replacement per finding, and a real null result when there is nothing to cut.
git
The pre-finish status: branch, hygiene findings, message checks, workflow lint, template state.
extend
Capture a mid-PR sub-idea and implement it onto the current open PR's branch — no new branch, no new PR. Preserves Verify → Review → Deliver.