You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Turn vague ideas into developer-ready specs and vertical-slice build plans — one question at a time. A gated, resumable interview that adapts its depth to project complexity. It never invents requirements: unknowns become traceable Open Questions tied directly to affected build slices. Delivered as a Claude Code skill.
Spec-driven governance for AI coding agents, organized in seven pillars (M.A.N.D.A.T.E.): Markdown as source of truth, Architecture by ADR, Named rules enforced by script, Decision Log, Agents with bounded mandates, Tested before asserted, Executable backlog.
Spec-driven development for codebases written with AI coding agents: a real SRS (ISO 29148, EARS, MADR) in plain Markdown, immutable requirement IDs, a generated traceability matrix, and a stdlib-only checker that fails the build when code and spec disagree. Works with Claude Code, Cursor, Codex, Copilot — or by hand.
An open source framework for AI-first engineering teams, born at Disruptica. Requests over roadmaps, contracts over tickets, agents that write the code. Companion tool: BugLady.ai / Ramjet