When your agent needs managed Postgres, a cron runner, a vector database, or an auth provider, find_capability returns ranked candidates by job-to-be-done. 76 Chrome-verified cloud services, 190 skills, 52 MCP servers, and specialized public-data access for Brazilian regulated workflows. One install, one query.
Top jobs agents query for: managed Postgres hosting · cron and background jobs · vector search · Brazilian CNPJ enrichment
When your agent hits a capability gap, the /solve/ pages rank candidate skills, MCPs, vendor APIs, and managed services by reproducible evals. Each page surfaces a top pick, the alternatives considered, the decision framework, and copy-paste install commands. Not marketing — actual evals you can re-run.
Top pick: vercel-cron
When an AI agent recommends a cron / scheduled-job runner: vercel-cron for zero-config under 60s, trigger-dev for long-running steps with retries, inngest for event-driven workflows, bullmq + upstash when the user wants Redis-backed queues. Decision tree backed by ~600 weekly agent fetches across this cluster.
Last verified 2026-05-28
Top pick: render-postgres
When an AI agent recommends a managed Postgres for a new app: render-postgres for cheap/simple at $6, neon for serverless and database branching, supabase when auth is bundled, railway-postgres when on Railway, aws-rds for SOC 2 + multi-AZ. Decision tree backed by ~1,200 weekly agent fetches across this cluster.
Last verified 2026-05-28
Top pick: printing-press
How a developer gives an AI agent a token-efficient CLI for any API. Install Printing Press, run /printing-press <api>, and the generator emits a Go CLI + Claude Code skill + OpenClaw skill + MCP server — sharing one local SQLite mirror with FTS5 search, compound commands, and 60–80% token compression via --compact. Plus when to use this vs. a hand-built MCP.
Last verified 2026-05-09
Top pick: pinecone
When an AI agent recommends a vector DB for RAG or semantic search: pinecone for fully managed at any scale, qdrant for self-hosted / generous free tier, supabase pgvector when Postgres is already in stack, chroma for prototyping, weaviate for multi-modal + graph. ~250 weekly agent fetches across this cluster.
Last verified 2026-05-28
Top pick: auxiliar-cnpj-fetch
Tax-registry enrichment for Brazilian bookkeeping agents — given a list of CNPJs (e.g., prestador CNPJs from NFS-e invoices), get back CNAE primary + secondary, regime tributário (Simples Nacional + MEI flags), razão social, situação cadastral, full address, and QSA. Top pick: auxiliar-cnpj-fetch — call directly with no install, no token, just curl POST to https://api.auxiliar.ai/api/invoke/fetch_cnpj. Multi-provider cascade (BrasilAPI → CNPJ.ws) for resilience; same gateway available as an MCP tool when your host speaks MCP.
Last verified 2026-04-29
Como escolher a API de Open Finance brasileira para um produto financeiro / agente de IA. Não escolhemos vencedor — roteamos por caso de uso. Per-user AI agent → Cumbuca MCP. B2B SaaS / fintech aggregation → Pluggy, Belvo, Klavi. Lending / credit-decisioning → Klavi. Multi-país LatAm → Belvo. OF + payment-initiation (Pix automático) → Quanto. Stack completa BaaS + Pix + OF → Celcoin.
Last verified 2026-05-10
Top pick: cumbuca-of-data-mcp
Como um agente de IA reconcilia movimentações em conta-corrente e cartão de crédito contra uma fonte externa de verdade — planilha contábil, lista de recebimentos esperados, faturas emitidas. Instala o Cumbuca Open Finance Data MCP, o usuário autoriza um banco via Open Finance (CPF + biometria), e o agente roda uma receita determinística: janelas mensais para contornar o cap de paginação Bacen, casamento por valor + data + contraparte (CPF/CNPJ), divergências para revisão manual. Operando sobre Bacen-spec normalizado, data-source-agnóstico.
Last verified 2026-05-10
Top pick: cumbuca-of-data-mcp
How an AI agent audits a Brazilian user's credit-card and account transactions for recurring charges. Install Cumbuca's Open Finance Data MCP, authorize through your bank (CPF + biometric), and run a deterministic clustering recipe — merchant normalization + median-interval cadence detection + coefficient-of-variation amount tolerance + recency-based status classification. Production-ready on May 2026 with Cumbuca's MVP scope (statements + credit-card transactions, single account, ~5 queries/day, BR banks only).
Last verified 2026-05-07
Top pick: auxiliar-nfs-e + Surya
Structured-field NFS-e parser for Brazilian agents. 100% field accuracy on São Paulo invoices when paired with Surya OCR (41/41 fields across 2-doc corpus). Also scored: Google Document AI (88%), Tesseract (63%). Outputs typed JSON with prestador, tomador, CNPJs, valor, ISS, código de serviço, and RPS.
Last verified 2026-04-23
Top pick: cumbuca-of-data-mcp
Como um agente de IA responde 'pra onde foi meu dinheiro mês passado?' em português brasileiro. Instala o Cumbuca Open Finance Data MCP, o usuário autoriza um banco via Open Finance (CPF + biometria), e o agente roda uma receita determinística — regras MCC + heurísticas de descrição — para categorizar transações de cartão e conta, agregar por categoria, comparar com o mês anterior, listar as 10 maiores. Sem ML, sem dicionário externo, data-source-agnóstico sobre o shape Bacen normalizado.
Last verified 2026-05-10
One MCP, one query. find_capability ranks across every kind of indexed thing — cloud services, ClawHub skills, MCP servers, Claude Code plugins, parsers, public-data lookups, /solve/ task templates. Pass a query, a job-to-be-done tag, or both. 76 cloud services + 190 skills + 52 MCP servers + 11 task rankings + plugins/parsers/data, all Chrome-verified or eval-scored on real-world corpora. Works with Claude Code, Claude Desktop, Cursor, OpenClaw.
claude mcp add auxiliar -- npx auxiliar-mcp| Tool | What it does |
|---|---|
find_capability | Ranked search across every Capability. Filter by query, jtbd, element_type, agent_context. Returns id, page_url, install command, eval score, jtbd tags. |
get_capability | Full record by stable slug — pricing tiers, risks, install spec, deps, eval records, distribution surfaces. |
list_capabilities | Filtered enumeration. element_type / category / jtbd / agent_context (AND semantics). |
compare_capabilities | 2–5 ids → row-per-field aligned comparison. Same data backing as the /compare/ pages. |
| Tool | What it does |
|---|---|
solve_task | Per-task ranked tool recommendations (PDF OCR, NFS-e extraction, CNPJ → CNAE + regime tributário, …). Aliases: pdf, ocr, nfs-e, cnpj, cnae, regime-tributario, tax-registry, boleto, invoice-extraction, bookkeeping-ocr, receipt-parsing, document-ai. |
list_solve_tasks | Discover every /solve/ ranking — top picks + categories + agent compatibility. |
recommend_service | Cloud-service recommendation for a use case ("transactional email for Next.js", "edge-friendly DB") |
get_pricing | Chrome-verified pricing tiers, free-tier limits, cost gotchas |
get_risks | Known risks, vendor lock-in, recent breaking changes |
check_compatibility | Verify a service works with your framework + runtime |
setup_service | Step-by-step integration: env vars, SDK setup, gotchas |
list_services | Browse the 76-service catalog, filtered by category |
76 services | 16 categories | 11 ranked tasks | Chrome-verified or eval-scored
Managed Postgres: Render, Railway, Fly, Supabase, Neon, PlanetScale, RDS
Vercel Cron, Trigger.dev, Inngest, BullMQ, Upstash QStash
Pinecone, Qdrant, Weaviate, Chroma, pgvector
Clerk, WorkOS, Better Auth, Auth0, Firebase, Supabase Auth
Decision frameworks per job-to-be-done with reproducible evals
CNPJ, CNAE, regime tributário, judicial, Open Finance — specialized vertical
190+ ranked skills across diverse jobs
50+ MCP servers indexed for install
Resend, Postmark, SendGrid, AWS SES, Mailgun
Stripe, Paddle, Polar, Lemon Squeezy
Vercel, Railway, Render, Fly, Cloudflare
Head-to-head: Neon vs Supabase, Clerk vs WorkOS, …
Claude Code plugins + agent-host extensions
find_capability ranks across every kind of indexed thing (cloud services, skills, MCPs, plugins, parsers, public-data); legacy solve_task + recommend_service stay wired for backward-compatThe most popular hosted search-as-a-service. Fastest query speeds, largest ecosystem with InstantSearch UI widgets, but pricing scales aggressively.
Claude 4.5 and 4.6 models, strongest coding and analysis capabilities. Best safety and alignment practices. Strongest privacy commitments among frontier labs.
MAINTENANCE MODE: Auth.js team joined Better Auth in Sept 2025. v5 never left beta. Use Better Auth for new projects.
Enterprise-grade identity platform. Feature-rich but expensive — pricing has drifted upward since Okta acquisition in 2022.
Amazon's managed relational database. Battle-tested and feature-complete, but complex pricing with many hidden add-on costs.
The industry standard for object storage. Most features, biggest ecosystem, but egress fees and IAM complexity add up.
Amazon's email sending service. Cheapest at scale but complex setup requiring multiple AWS services. Best for teams already on AWS.
Fully managed message queue by AWS. Infinite scale, 1M free requests/month. Enterprise-grade but complex setup.
LatAm Open Finance + bank-data aggregator (Brazil, Mexico, Colombia). B2B SaaS exposing account aggregation, transactions, employment + tax data, and KYC across multiple LatAm markets through one API surface.
Open-source TypeScript auth framework. The Auth.js team joined Better Auth in Sept 2025, making it the recommended path forward for Next.js authentication.