An agentic coding workflow for product builders
SPECTRE is an agentic Coding Workflow - /Scope, /Plan, /Execute, /Clean, /Test, /Rebase, /Evaluate - that uses simply step by step product development workflow to generate high quality results from your AI Coding Agents. - Codename-Inc/spectre
SPECTRE enables 10-100x output acceleration by combining rapid waterfall (specificity upfront) with iterative refinement, allowing developers to reach working code faster and iterate on features rather than debugging ambiguous implementations.
SPECTRE's /spectre:handoff command preserves session state and work threads in managed documentation blocks, enabling agents to resume work with full context awareness and reducing the need for manual context re-entry between sessions.
SPECTRE delivers a single unified workflow that adapts to brand-new codebases and large existing projects (tested on 250k+ line codebases), eliminating the need to switch processes between small and large features while maintaining engineering rigor.
SPECTRE provides a structured slash-command workflow (Scope → Plan → Execute → Clean → Test → Rebase → Evaluate) that generates canonical documentation and reduces ambiguity upfront, enabling AI agents to produce higher-quality, more consistent code with longer autonomous execution periods.
SPECTRE automatically creates canonical docs (scope.md, ux.md, plan.md, tasks.md, code_review.md, skills) stored in docs/tasks/{topic}/specs, providing a structured knowledge base that agents can reference and teams can use for future iterations.