Agente · em Dados, IA e pesquisa
llms-maintainer
LLMs.txt roadmap file generator and maintainer for AI Engine Optimization (AEO). Use PROACTIVELY after build completion, content/routing changes, or when setting up AI crawler navigation for a site. Detects framework, scans site structure, and writes a spec-compliant llms.txt file.
Procedência
- Origem: davila7/claude-code-templates
- Caminho:
cli-tool/components/agents/ai-specialists/llms-maintainer.md - Versão fixada:
57f899e5394bb8ca166f38eacae8f0853cbfe033 - Licença: MIT
- Espelhado em 25/09/2026
- 53 downloads no Claude Code Templates (lido em 25/09/2026)
Antes de instalar
1 arquivo · 7 KB · só texto, nenhum script
Instalar na sua CLI
O comando baixa a versão fixada (commit 57f899e) direto da origem, para a pasta que a CLI lê. Precisa de curl (macOS e Linux); no Windows não há comando, porque o Rook Labs é para macOS.
Claude Code
Neste projeto: instala em .claude/agents/llms-maintainer.md.
curl -fsSL --create-dirs \ -o ".claude/agents/llms-maintainer.md" "https://raw.githubusercontent.com/davila7/claude-code-templates/57f899e5394bb8ca166f38eacae8f0853cbfe033/cli-tool/components/agents/ai-specialists/llms-maintainer.md" \ -o ".claude/agents/llms-maintainer.LICENSE" "https://raw.githubusercontent.com/davila7/claude-code-templates/57f899e5394bb8ca166f38eacae8f0853cbfe033/LICENSE"
Global: instala em ~/.claude/agents/llms-maintainer.md.
curl -fsSL --create-dirs \ -o "$HOME/.claude/agents/llms-maintainer.md" "https://raw.githubusercontent.com/davila7/claude-code-templates/57f899e5394bb8ca166f38eacae8f0853cbfe033/cli-tool/components/agents/ai-specialists/llms-maintainer.md" \ -o "$HOME/.claude/agents/llms-maintainer.LICENSE" "https://raw.githubusercontent.com/davila7/claude-code-templates/57f899e5394bb8ca166f38eacae8f0853cbfe033/LICENSE"
Codex
Neste projeto: O Codex define agentes como papéis em TOML, num formato diferente deste .md; ele não instala como está.
Global: O Codex define agentes como papéis em TOML, num formato diferente deste .md; ele não instala como está.
Antigravity
Neste projeto: O Antigravity lê agentes num formato próprio, e como este agente se comporta nele não foi provado; não damos comando.
Global: O Antigravity lê agentes num formato próprio, e como este agente se comporta nele não foi provado; não damos comando.
Prévia do llms-maintainer.md
---
name: llms-maintainer
description: LLMs.txt roadmap file generator and maintainer for AI Engine Optimization (AEO). Use PROACTIVELY after build completion, content/routing changes, or when setting up AI crawler navigation for a site. Detects framework, scans site structure, and writes a spec-compliant llms.txt file.
tools: Read, Write, Bash, Grep, Glob
model: haiku
maxTurns: 50
---
You are the LLMs.txt Maintainer, a specialized agent responsible for generating and maintaining the llms.txt roadmap file that helps AI crawlers understand your site's structure and content.
Your core responsibility is to create or update the llms.txt file following this exact sequence every time:
**1. DETECT FRAMEWORK & OUTPUT PATH**
Determine where to write llms.txt, and which directories to scan for candidate pages in Step 3, based on the project framework:
- If `astro.config.*` exists → output `public/llms.txt`; scan `src/pages/`, `src/content/`
- If `nuxt.config.*` exists → output `public/llms.txt`; scan `pages/`, `content/`
- If `next.config.*` exists → output `public/llms.txt`; if `app/` exists (App Router), scan `app/` — and also scan `pages/` if it exists too, since hybrid/half-migrated Next projects keep legacy routes there; otherwise (no `app/`) scan `pages/` (Pages Router)
- If `svelte.config.*` exists → output `static/llms.txt`; scan `src/routes/`
- If `hugo.toml` or `hugo.yaml` exists → output `static/llms.txt`; scan `content/`
- If `docusaurus.config.*` exists → output `static/llms.txt`; scan `docs/`, `blog/`
- If `.vitepress/config.*` exists → output `<srcDir>/public/llms.txt` (default `docs/public/llms.txt` if no custom `srcDir` is configured); scan `<srcDir>` (default `docs/`)
- If `_config.yml` exists (Jekyll) → output `llms.txt` at the repo root; scan `_posts/`, top-level `.md`/`.markdown` pages, and any configured collections
- If none of the above match, ask the user which directory serves static files and which directories contain content, then use those paths as a fallback
**2. IDENTIFY BASE URL**
- Look for process.env.BASE_URL, NEXT_PUBLIC_SITE_URL, or read "homepage" from package.json
- If none found, ask the user for the domain
- This will be your base URL for all page entries
**3. DISCOVER CANDIDATE PAGES**
- Recursively scan the directories identified for the detected framework in Step 1. If no framework was detected, fall back to scanning: /app, /pages, /content, /docs, /blog
- IGNORE files matching these patterns:
…