authenticfake/clike

🚀 CLike — AI-Native Pipeline for Product Engineers

From intent to impact.
CLike keeps developers in flow, augments delivery with agentic workflows, and bakes in governance, eval-driven quality, repository grounding, and a safe paved road for enterprise-grade software delivery.

---

🚀 Project Status & Release Info

| 📅 Last Updated | 🧪 Current Testing | 📦 Latest Release | | --- | --- | --- | | June 9, 2026 | Agent execution reach expanded, in chronological order: (1) Harper phases via execution-agent extended beyond /kit, /eval, and /finalize to also cover /idea, /spec, and /plan; (2) CLike skills strengthened to make the local agent effective across the /plan, /kit, /eval, and /finalize Harper phases; (3) Free chat (Q&A) and Coding now run via local agent (Claude Code / Codex CLI) in addition to the cloud path, across all CLike chat modes. Local checks: python -m py_compile orchestrator/routes/v1.py gateway/routes/models.py gateway/providers_availability.py; PYTHONPATH=orchestrator:. pytest -q orchestrator/tests; node --test extensions/vscode/test/*.test.js. | | | June 2, 2026 | BMAD-aware methodology profile MVP documented and under regression testing: governed methodology context, companion artifacts, SPEC PM/UX safety, server-side companion discovery, cloud prompt verification, local-agent package verification, eval/gate boundaries, and deterministic scorecards. Real local checks: python -m compileall orchestrator gateway; `PYTHONPATH=orchestrator:. pytest -q orchestrator/tests/test_methodology_resolver.