AGENA — Agentic AI Platform for Autonomous Code Generation
A multi-tenant agentic AI platform that coordinates LLM-powered agents to analyze tasks, generate code, review changes and open pull requests autonomously, running on an existing Claude Code or Codex CLI subscription instead of metered API billing.
AGENA is a production-oriented, multi-tenant agentic AI platform, built as a monorepo of six pip-installable packages, that runs an autonomous PM, planner, developer, reviewer and finalizer pipeline: it analyzes a task, generates code, reviews the change, and opens a pull request without further intervention. Rather than billing per token, it runs the actual work through a host bridge to an existing Claude Code or Codex CLI subscription.
It closes the loop from a production error captured in Sentry or New Relic through to a task, then an AI fix, then a pull request, then a merged and auto-resolved issue, and it pulls real work items from Sentry, New Relic, Jira, YouTrack, Azure DevOps or GitHub. A history-grounded sprint-refinement feature indexes closed Azure DevOps or Jira work items, including their final story points, assignees and pull request details, into a per-organization vector collection, so a new item's story-point suggestion is anchored on the most similar past items. CrewAI role-based agents and a LangGraph state machine drive the pipeline, a Prompt Studio lets system prompts be edited at runtime without a code deployment, and vector memory in Qdrant lets the system learn from previous tasks.
It is self-hostable and multi-tenant, keeping code and repositories on the operator's own infrastructure.