-
coreci-bot released this
2026-09-11 16:16:51 +00:00 | 199 commits to main since this releaseNextcraft v0.1.5 — v0.2 Phase 4
The build-and-grade agents join the roster — real LLM logic over mock engine inputs (simulated telemetry, pre-baked artifacts; real engines are v0.3+, per D-015).
Delivered (REQ-2-007, REQ-2-008)
- Lab agent (persona lab-v2) — streams concrete in-flow feedback from sandbox telemetry: what the events show, the one most useful adjustment, one next step. Coaching tone on suspicious patterns (huge paste → self-check suggestion, not accusation)
- Assessor agent (persona assessor-v2) — applies rubrics to artifacts + defense transcripts; returns pydantic-validated structured scores with cited evidence, strengths, gaps, verdict; the 4-layer JSON defense (response_format degrade → prompt schema → defensive parse → bounded retry) is fully exercised
- Endpoints —
POST /v1/lab/feedback(SSE, D-016 envelope, 404 unknown scenario);POST /v1/assessment/evaluate(validated JSON, 404/422/502 paths) - Mock input corpora — 3 telemetry scenarios (strong/struggling/flagged), 2 pre-baked artifacts with defense transcripts, 2 rubrics; IDs string-identical on the TS and Python sides (D-021, mechanically cross-checked)
Quality: 110/110 tests green (cloud-free), ruff clean, tsc green, lockfile integrity restored (apps/ai-service importer).
Next
Phase 5: Proctor + Mentor agents — integrity signals with coaching interventions, long-horizon career narrative.
Downloads
-
2026-09-11 15:59:57 +00:00 | 200 commits to main since this releaseNextcraft v0.1.4 — v0.2 Phase 3
The first two fully-implemented AI tutor agents, live and streaming.
Delivered (REQ-2-005, REQ-2-006)
- Coach agent — pacing guidance, motivation tied to trajectory, retrieval-practice questions; always ends with exactly one concrete next action (persona coach-v2)
- Tutor agent — one concept per reply, worked example first, Socratic check question before advancing (persona tutor-v2)
- Agent routing —
/v1/chat/streamroutes by agent name; unknown agents rejected 422; meta event carries the routed agent; sessions scoped per agent (no persona bleed) - Live-verified personas — manual ollama-cloud probe (gemma4:31b): Coach responded with pacing + retrieval question + one action; Tutor with concept + worked example + Socratic question. Distinct voices, on-persona.
Quality: 80/80 tests green (mock provider only — suite never calls the cloud), ruff clean, boundaries verified.
Next
Phase 4: Lab + Assessor agents over mock engine inputs (simulated telemetry, pre-baked artifacts).
Downloads
-
2026-09-11 15:52:00 +00:00 | 201 commits to main since this releaseNextcraft v0.1.3 — v0.2 Phase 2
The shared framework all six AI tutor agents build on.
Delivered (REQ-2-004)
- BaseAgent ABC — system_prompt / build_messages / stream_reply / structured_reply; default provider delegation
- Session store — protocol + in-memory implementation: asyncio.Lock-guarded, 20-message rolling window, 500-cap LRU eviction, agent-scoped sessions (DB-migration-ready)
- Agent registry — central registration, duplicate/unknown error paths
- Structured output defense — 4 layers: response_format with auto-degrade, prompt-embedded schema, fence-strip/first-balanced-object parse, single bounded retry
- Prompt library — six versioned modules with learner-context renderers
- Learner context corpus — pydantic-typed, IDs string-identical to the TS mock-data layer
- Chat endpoint session wiring — persisted turn history, windowed replay, assistant-turn persistence
Quality: 61/61 tests green, ruff clean, module boundaries verified (agents/ never imports api/).
Next
Phase 3: Coach + Tutor agents — full LLM implementations with agent routing.
Downloads
-
2026-09-11 15:40:30 +00:00 | 202 commits to main since this releaseNextcraft v0.1.2 — v0.2 Phase 1
apps/ai-servicestands up: a Python FastAPI application hosting the AI tutor agents behind a provider-agnostic LLM layer.Delivered (REQ-2-001, REQ-2-002, REQ-2-003)
- FastAPI app — lifespan-managed httpx pool (10s connect / 300s read for cloud TTFT), localhost-only CORS,
/health - Provider-agnostic LLM layer —
LLMProviderprotocol;OpenAICompatProvider(ollama-cloud + local endpoints, raw httpx, no SDK); deterministicMockProvider(failure injection, abort recording); factory selection - SSE chat streaming —
POST /v1/chat/streamwith the D-016 envelope:metaevent → content deltas →done→[DONE]; in-banderrorevents on failure - Monorepo integration — zero-dep package shim, turbo passthrough tasks, root
pnpm ai:dev/ai:test/ai:lint/ai:bootstrap - Quality gates — ruff clean, 20/20 pytest green (mock provider only, mechanically enforced cloud-free guard), live smoke verified meta→delta→done→[DONE]
Next
Phase 2: agent framework (BaseAgent, session store, registry, structured outputs).
Downloads
- FastAPI app — lifespan-managed httpx pool (10s connect / 300s read for cloud TTFT), localhost-only CORS,
-
v0.1.1 — v0.2 Pre-Execution Stable
2026-09-11 15:16:02 +00:00 | 204 commits to main since this releaseNextcraft v0.1.1 — v0.2 Pre-Execution (Phase 0)
Milestone v0.2 (AI Tutor Architecture) phase 0 complete. This patch ships the full specification, research, and plan for the six AI tutor agents (Coach, Tutor, Lab, Assessor, Proctor, Mentor) as a Python FastAPI
apps/ai-service.What's in this release
- SPECIFY — REQ-2-001..012 activated; decisions D-013 (v0.1 founder-agreed, business logic unlocked), D-014 (provider-agnostic LLM, ollama-cloud default), D-015 (all six agents, engines mocked)
- CLARIFY — 10 ambiguities auto-resolved (A-001..A-010)
- RESEARCH — FastAPI 0.141 + pydantic v2 + httpx + sse-starlette stack; D-016..D-023 design decisions; ai-engineer persona added, backend-engineer reactivated
- PLAN — 6 execution phases, 39 tasks, 18 waves, persona-assigned with must-have checklists
- GRILL — adversarial review: GO-WITH-CHANGES (0.82); binding decisions G-1..G-5 applied
- MVP/UX gate — passed (User-Facing Surface, Happy Path, UX Acceptance Criteria)
Next
Execution phases P1-P6 (apps/ai-service scaffolding → agent framework → Coach+Tutor → Lab+Assessor → Proctor+Mentor → learner surface integration), shipping as v0.1.2..v0.1.7. Milestone release: v0.2.0.
Downloads
-
2026-09-10 22:48:00 +00:00 | 207 commits to main since this releaseNextcraft v0.1.0 — UI/UX Prototype Milestone Release
What This Is
High-fidelity interactive prototype of all 4 Nextcraft surfaces with realistic mock data, navigation flows, responsive layouts, and a shared component library. No backend, no business logic — everything static/mocked.
Surfaces (4)
Learner Surface (7 pages)
- Landing page with hero, Byte→Build→Demonstrate→Defend flow, program highlights, testimonials
- Program catalog with 5 competency stacks (70 total competencies)
- Competency stack view with status indicators, progress bars, microcredential badges
- Learner dashboard with 6 panels including AI tutor chat mockup and recharts progress graph
- Byte tutorial viewer with concept panel and tabbed worked examples
- Build sandbox mockup with IDE layout (toolbar, file explorer, editor, telemetry)
- Assessment/defense mockup with rubric, AI reviewer, oral defense, process trace timeline
Marketplace Surface (5 pages)
- Job board with 20 mock AI-era job listings, match scores, skill tags
- Client-side search and filter (skills, seniority, remote, salary range)
- Job detail with competencies, AI-matched skills breakdown, employer info, related jobs
- Employer profiles with culture section and open positions
- Pricing page with 6 tiers and feature comparison table
Employer Dashboard (4 pages)
- Dashboard overview: 4 metric cards, applicant pipeline (5 columns), 3 recharts analytics charts
- Talent search: 15 candidate cards with client-side filtering
- Candidate profiles: artifact gallery, process trace, defense transcripts, microcredential verification
- Posting management: posting list + detail form + applicant table
Admin Surface (4 pages)
- Admin overview: 5 platform metrics, activity feed (10 events), system health (6 services)
- Learner management: searchable table with detail panel (15 mock learners)
- Competency graph viewer: interactive @xyflow/react graph (20 nodes, clickable, minimap)
- Marketplace moderation: 3 tabbed queues with action buttons
Tech Stack
- Next.js 15 (App Router, Server Components)
- Tailwind CSS v4 (@theme, dark mode via @custom-variant)
- TypeScript 5.x (strict)
- pnpm 12 + turborepo (monorepo)
- lucide-react (icons)
- recharts (charts)
- @xyflow/react (competency graph viewer)
- Inter font (next/font)
- Storybook 10 (6 stories)
Structure
nextcraft/ apps/web/ # Next.js app (19 routes) packages/ui/ # Shared component library (5 primitives + tokens) packages/types/ # Shared TypeScript types packages/mock-data/ # Typed mock data (5 stacks, 20 jobs, 15 candidates, 10 employers)Requirements: 28/28 complete
- Infrastructure: REQ-001 to REQ-005 ✅
- Learner Surface: REQ-006 to REQ-012 ✅
- Marketplace Surface: REQ-013 to REQ-017 ✅
- Employer Dashboard: REQ-018 to REQ-021 ✅
- Admin Surface: REQ-022 to REQ-025 ✅
- Polish & Integration: REQ-026 to REQ-028 ✅
Phases (7)
- P0 Pre-execution → v0.0.1
- P1 Project Scaffolding → v0.0.2
- P2 Learner Surface → v0.0.3
- P3 Marketplace Surface → v0.0.4
- P4 Employer Dashboard → v0.0.5
- P5 Admin Surface → v0.0.6
- P6 Polish + Integration → v0.0.7
- P7 Final Review + Ship → v0.1.0 (this release)
Verification
- ✅ pnpm build: 4/4 packages, 19 routes
- ✅ pnpm typecheck: 7/7 tasks
- ✅ storybook build: all stories compile
- ✅ 104 TypeScript files, 19,536 lines of code
- ✅ 26 decisions logged (D-001 through D-026)
- ✅ Dark mode on all surfaces
- ✅ Breadcrumbs on all 20 routes
- ✅ Responsive (375px / 768px / 1280px)
- ✅ WCAG AA contrast verified
Downloads
-
2026-09-10 21:31:49 +00:00 | 207 commits to main since this releaseNextcraft v0.0.1 — Pre-execution Phase
Completed Stages
- SPECIFY: Specification validated against PROJECT.md and REQUIREMENTS.md (28 REQ-IDs)
- CLARIFY: 8 ambiguities auto-resolved (D-013 through D-020)
- RESEARCH: Tech stack confirmed — Next.js, Tailwind CSS v4, lucide-react, recharts, @xyflow/react. Personas assessed.
- PLAN: 6 execution phases, 20 tasks, wave-ordered with persona assignments
- MVP/UX CHECK: Passed — user-facing surface, 14-step happy path, 12 acceptance criteria
Files Created
- .ciagent/config.json
- .ciagent/PROJECT.md
- .ciagent/REQUIREMENTS.md
- .ciagent/ARCHITECTURE.md
- .ciagent/ROADMAP.md
- .ciagent/PLAN.md
- .ciagent/PERSONAS.md
- .ciagent/CHECKPOINT.json
- .gitignore
Decisions
26 decisions logged (D-001 through D-026)
Next Steps
Phase 1: Project Scaffolding — monorepo setup, component library, mock data, routing
Downloads