AgentEra/Agently

[GenAI Application Development Framework] 🚀 Build GenAI application quick and easy 💬 Easy to interact with GenAI agent in code using structure data and chained-calls syntax 🧩 Use Event-Driven Flow *TriggerFlow* to manage complex GenAI working logic 🔀 Switch to any model without rewrite application code

What it solves

Agently is an AI application runtime framework designed to move AI services from experimental prompts to reliable production backends. It solves the problem of "output drift" and instability by providing stable contracts between the model and the application code, ensuring that model changes or provider swaps do not break business logic.

How it works

Agently acts as an execution substrate that normalizes model providers, prompt slots, and response parsing into a single request/runtime contract. It uses a "contract-first" approach to structured output, where schemas are defined as prompts and validated by the framework. It also features TriggerFlow for signal-driven orchestration, allowing workflows to react to partial structured output (via "instant" mode) or action results without waiting for a full model response.

Who it’s for

  • Product Engineers building AI assistants, internal copilots, or AI-backed APIs who need reliability and maintainability.
  • Platform Teams requiring clear extension points for model providers, MCP servers, and observability.
  • Technical Leads looking for a framework that prioritizes explicit control and debuggability over broad, prebuilt agent integrations.

Highlights

  • Structured Output Guarantees: Framework-level validation, required-field extraction, and retries that work independently of provider-native features.
  • Instant Streaming: Ability to react to specific structured fields while the model is still streaming tokens.
  • Observable Actions: Support for local functions, MCP servers, and managed execution resources (Python, Shell, SQLite) with structured logs and artifact references.
  • Signal-Driven Workflows: TriggerFlow supports event-driven orchestration, pause/resume capabilities, and state snapshots for recoverable workflows.
  • Model Agnostic: Normalizes provider setup and response parsing to make switching models seamless.