Factorly
factorly-dev/factorly
Runtime seguro para orquestração de tool chains com cofre de credenciais
Ver no GitHubRuntime local para orquestração de tool chains de agentes de IA. Gerencia chamadas de ferramentas, injeta credenciais de um cofre criptografado, aplica regras de governança e registra tudo. Mantém segredos realmente secretos enquanto conecta agentes a dezenas de ferramentas.
Instalação
npx convez-ai factorly-dev/factorly1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
Casos de uso
- Integrar agentes de IA com múltiplas ferramentas e APIs de forma segura
- Centralizar gerenciamento de credenciais e chaves com criptografia
- Auditar e governar todas as chamadas de ferramentas do agente
- Testar workflows de agentes antes de colocá-los em produção
Veredito

Solução robusta e bem-pensada para o problema crítico de credenciais em agentes. A arquitetura de injeção de credenciais em tempo de execução, cofre criptografado nativo e suporte multi-protocolo (MCP, REST, CLI) a tornam essencial para deployments sérios. Documentação clara e blueprints prontos aceleram a adoção.
README
A tradução para o português deste README ainda está na fila.
░█▀▀░█▀█░█▀▀░▀█▀░█▀█░█▀▄░█░░█░█
░█▀▀░█▀█░█░░░░█░░█░█░█▀▄░█░░░█░
░▀░░░▀░▀░▀▀▀░░▀░░▀▀▀░▀░▀░▀▀▀░▀░
Factorly
Build what your agent can do.
Define tools, compose workflows, test and run them. MCP servers, REST APIs, and CLI commands in one config, one UI, one audit log.
Factorly is a local runtime for agent tool chains. It manages tool calls, injects credentials from an encrypted vault, enforces governance rules, and logs everything. Your agent sees workflows, tools, and data. It never receives your credentials, only the tool's response. Secrets stay secret.

Install
Install with your package manager of choice:
brew install factorly-dev/tap/factorly
# or: npm install -g factorly
# or: pip install factorly
# or: go install github.com/factorly-dev/factorly@latest
Quick Start
Zero config: wrap any existing MCP server and it's instantly callable through Factorly:
factorly wrap -- npx -y @modelcontextprotocol/server-everything
Then set up a project: install a blueprint, store credentials, and sync with your agent:
# 1. Create a project config + install a blueprint (41 services: GitHub, Slack, Stripe, Linear, Gmail, ...)
factorly init
factorly blueprint install github
# 2. Store credentials in the encrypted vault (or: factorly auth login github)
factorly vault set GITHUB_TOKEN ghp_xxxxxxxxxxxx
# 3. Confirm it's wired up
factorly tools status
# 4. Connect to your agent (auto-detects Claude Code, Cursor, Codex), then reload it
factorly sync
# 5. Optional — browse, edit, and try tools in the UI
factorly ui
Your agent connects to Factorly as a single MCP server or CLI and sees every tool you've configured. Your agent never possesses the credential — Factorly resolves {{vault:KEY}} references at call time and only the resolved request leaves the box. Full walkthrough: Getting Started.
What It Does
Define — one config, every protocol, 40+ blueprints included
Test — Try tools in the UI, see the response, iterate before giving your agent access
Compose — workflows with per-step policies, deterministic sequences
Govern — vault, policies, audit log. Built in, not bolted on.
┌────────────┐ ┌────────────┐ ┌────────────┐
│ │ │ │ │ │
│ Your Agent │──────▶│ Factorly │──────▶│ Your Tools │
│ │ │ │ │ │
└────────────┘ └────────────┘ └────────────┘
Sees: Vault REST APIs
- tool names Governance CLI commands
- workflows Audit log MCP servers
- data Rate limits
Never sees:
- API keys
- tokens
- credentials
Docs
- Getting Started
- Configuration
- CLI Reference
- Web UI
- Workflows
- Expressions
- Output Filters
- Blueprints
- Vault
- OAuth
- Logging
- OpenAPI Import
- Examples