git:20260708.92097e7 to git:20260708.60d981f

1 added, 1 removed. Audit A to A.

# SDLC Studio

> An open Agent Skill (agentskills.io format) that runs the full software development

> lifecycle - PRD, epics, stories with executable acceptance criteria, implementation,

> testing, verification, reconciliation - with deterministic gates that hold the agent to

> what the files actually say. Installs into Claude Code, Codex, Gemini CLI, opencode and

> GitHub Copilot; invoked as /sdlc-studio [type] [action] or model-invoked from plain

> language.

## Docs

- [README](README.md): install, quick start, capability catalogue

- [Why SDLC Studio](docs/why-sdlc-studio.md): the value argument, evidence and economics

- [SKILL.md](.claude/skills/sdlc-studio/SKILL.md): the skill entry point and router

- [AGENTS.md](AGENTS.md): guidance for coding agents working on this repository

- [Install reference](docs/INSTALL.md): all tools, scopes and flags

## Evidence

- [Benchmark protocol](docs/benchmarks/protocol-v2.md): pre-registered measurement against plain AI coding

- - [Latest results](docs/benchmarks/2026-07-08-v2-respike.md): published regardless of outcome

+ - [Latest results](docs/benchmarks/2026-07-08-n5-run.md): the N=5 measured run, published regardless of outcome