hiteshbandhu Navigate AIE Europe 2026
Action playbook from fourteen AIE Europe 2026 keynotes. Do not summarize talks — pick a workflow and execute it.
Supporting files:
- workflows.md — workflows A–N
- source-index.md — src-NNN → learnings
Optional deliverables: {SKILL_OUTPUT_DIR}/navigate-aie-europe-2026/
Step 0 — Pick workflow
What is the user trying to do?
├─ Agent orchestration / AgentCraft → A [src-001]
├─ Agents beyond chat (legal/work UX) → B [src-002]
├─ Quality bar in slop era (pi / craft) → C [src-003, src-010]
├─ Code mode / code-as-interface → D [src-004]
├─ How AI changes engineering (Pragmatic Engineer) → E [src-005]
├─ Gemini / DeepMind research direction → F [src-006]
├─ Replace large code with small skills → G [src-007]
├─ Software fundamentals still matter → H [src-008]
├─ OpenClaw / local agent state → I [src-009]
├─ End of apps / new UX containers → J [src-011]
├─ Judgment vs friction (human in loop) → K [src-012]
├─ MCP roadmap & agent stack → L [src-013]
├─ New application layer (Vercel) → M [src-014]
└─ Skills + subagents in products → G [src-007] + [L]
Open workflows.md.
Install
cp -r skills/navigate-aie-europe-2026 ~/.claude/skills/
cp -r skills/navigate-aie-europe-2026 ~/.cursor/skills/
Source: playlists/aie-europe-2026-keynotes/.
Cross-cutting rules
| Rule | Source |
|---|---|
| 2026 agents = skills + MCP + production hardening | [src-013 @ 5:31] |
| Load prompts/skills only when user invokes | [src-007 @ 8:39] |
| Code mode: let executable code carry intent | [src-004] |
| Judgment remains human; friction is signal | [src-012] |
| Fundamentals (types, tests) still gate quality | [src-008] |
Output to user
- Name workflow (A–N) and deliverable
- Artifacts under
./skill-outputs/navigate-aie-europe-2026/when requested
Invocation examples
@navigate-aie-europe-2026 MCP adoption plan for our agent
should we replace our rules engine with a 200-line skill?
code mode vs tool-calling for our workflow engine