Kanevry/session-orchestrator

48 stars · Last commit 2026-05-30

Turn ad-hoc Claude Code / Codex / Cursor sessions into a repeatable loop — research → plan → wave-execute → close. Parallel sub-agents, inter-wave quality gates, dual GitHub+GitLab integration, F2 Memory + Persona-Panel + gsd-Pattern Adoption. 37 skills · 18 commands · 13 agents · 7360 tests.

README preview

# Session Orchestrator

[![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](LICENSE)
[![Version](https://img.shields.io/badge/version-3.8.0-blue.svg)](CHANGELOG.md)
[![Tests](https://img.shields.io/badge/tests-7360%20passing-brightgreen.svg)](#development)
[![Claude Code](https://img.shields.io/badge/Claude_Code-Plugin-blueviolet.svg)](https://docs.anthropic.com/en/docs/claude-code)
[![Codex](https://img.shields.io/badge/Codex-Compatible-green.svg)](https://developers.openai.com/codex/)
[![Cursor IDE](https://img.shields.io/badge/Cursor_IDE-Compatible-blue.svg)](https://cursor.com)

Turn ad-hoc Claude Code sessions into a repeatable loop with verification gates. Session Orchestrator adds a structured `research → plan → execute in waves → close` cycle on top of your existing agent — across Claude Code, Codex CLI, and Cursor IDE. Inter-wave reviews catch regressions before they ship; carryover issues mean nothing slips through.

## What you get

- **38 skills** for the session lifecycle (start, plan, execute, close, evolve), discovery, vault sync, MCP authoring, debugging, brainstorming, persona panels, harness/repo audits, tmux visualization, and more
- **18 slash commands** (`/session`, `/go`, `/close`, `/discovery`, `/plan`, `/evolve`, `/autopilot`, `/test`, `/brainstorm`, `/debug`, `/persona-panel`, `/memory-cleanup`, …)
- **13 typed sub-agents** (code-implementer, test-writer, security-reviewer, session-reviewer, qa-strategist, architect-reviewer, dialectic-deriver, memory-proposal-collector, …)
- **14 hook event handlers** enforcing scope, blocking destructive commands, gating templates-first, auditing memory proposals, capturing telemetry
- **7360 vitest tests** passing on every commit, validate-plugin 94/94, typecheck 230 files OK, lint 0

## Lifecycle at a glance

View full repository on GitHub →