[FEATURE] Modes, contexts & persona system — composable UX layers (Serena-inspired)

[Wolfvin]

Summary

Add 3 composable UX layers from Serena: Modes (9 composable YAML fragments affecting command availability + system prompt), Contexts (14 per-AI-client configs auto-detected from env vars), Persona (junior-dev / PM / power-user adaptive output).

Worker consensus (2 reports)

Worker	Source	Contribution
Serena	update!/CodeLens_vs_Serena_Upgrade_Analysis.md S7	Modes system — 9 modes: onboarding, editing, planning, interactive, one-shot, no-memories, no-onboarding, query-projects, mode.template. Composable: codelens mode onboarding,interactive. Affects command availability + system prompt. Custom mode at .codelens/modes/custom.yml.
Serena	same file S8	Contexts system — 14 contexts: claude-code, codex, cursor, vscode, ide, copilot-cli, chatgpt, antigravity, jb-ai-assistant, jb-copilot-plugin, junie, agent, oaicompat-agent, desktop-app. Auto-detect from env vars (CLAUDE_CODE=true, CURSOR=true, etc.). Custom context at .codelens/contexts/custom.yml.
UnderstandAnything	update!/CodeLens_vs_UnderstandAnything_Upgrade_Analysis.md U11	Persona-adaptive output — 3 personas: junior-dev (verbose, more context, simpler language, include languageLesson), PM (high-level summary, business impact, less technical detail, include domain mapping), power-user (concise, technical, complexity/edge count/blast radius). `--persona <junior-dev

Proposed scope (P3, 3-5 weeks total — all optional)

Phase 1 — Modes system (P3, 1-2 weeks, depends on config system issue)

• 9 built-in modes at scripts/modes/*.yml
• Each mode has: description, prompt, excluded_commands, included_optional_commands
• Commands: codelens mode <name>, codelens mode list, codelens mode show <name>
• Composable: codelens mode onboarding,interactive
• Custom mode at .codelens/modes/custom.yml
• MCP tools: set_mode, list_modes, get_mode

Phase 2 — Contexts system (P3, 2 weeks, depends on config system issue)

• 14 built-in contexts at scripts/contexts/*.yml
• Auto-detect from env vars
• Commands: codelens context <name>, codelens context list
• MCP tools: set_context, list_contexts, get_context
• Custom context at .codelens/contexts/custom.yml

Phase 3 — Persona-adaptive output (P3, 1 week)

• --persona <junior-dev|pm|power-user> flag for summary, handbook, tour, explain
• Config field persona for default
• MCP tool exposes persona parameter
• Persona-specific prompt templates at scripts/prompts/
• Update summary_engine.py, handbook command, tour_generator.py

Acceptance criteria

• Modes affect command availability (e.g. no-memories mode hides memory commands)
• Contexts auto-detect from env vars (Claude Code, Cursor, VS Code)
• Persona affects output verbosity and focus
• Custom modes/contexts can be loaded from .codelens/

Related

• All 3 phases depend on the configuration system overhaul issue (multi-layer YAML)
• Phase 3 (persona) depends on tour_generator from UA U6 (guided tour generation) — if not shipped, persona just affects summary and handbook