Lingtai-AI

Nested lingtai-dev-guide reference for known implementation footguns: Bubble Tea v2 paste, textarea theming, dev-mode rebuilds, editable installs, migrations, localization, authorization gates, and config conventions.

Router for contributing to the LingTai project. Use this when you are about to change LingTai code or docs, set up a dev environment, navigate the Go TUI/portal repo or Python kernel, develop MCP addons, prepare a release, troubleshoot a running network, audit security, govern avatars, prepare a publication-bound release workflow, run a runtime self-check, get a PR review-ready, or steward a new skill. This is for developers and contributors; for end-user lessons, use tutorial-guide.

Compact lingtai-dev-guide release overview: when you are doing a release, the maintainer-authorization boundary, the version scheme, and a pointer to release-workflow for the full TUI/Portal + kernel publishing checklist, GitHub/PyPI/Homebrew steps, the required HTML release log, and the website release blog.

Nested lingtai-dev-guide reference for consequential LingTai releases: paired TUI/Portal + kernel release planning, clean worktrees, validation gates, GitHub/PyPI/Homebrew publishing boundaries, the required self-contained HTML release log, website release-log/blog drafting, and the reusable release blog template.

Nested lingtai-dev-guide reference for getting a PR review-ready: PR readiness gates, multi-model/daemon/Claude read-only review passes, self-contained local HTML explainers, PR body hygiene and gh pr edit troubleshooting, source-labeled deliverables with syntax/validation checks, and maintainer authorization boundaries for opening, editing, and merging PRs.

Nested lingtai-dev-guide reference for developer/operator runtime self-checks after a refresh, checkout, or preset/MCP change: probe which lingtai code is actually running, confirm the editable source and git HEAD, verify the active TUI/portal binary and dev-mode symlinks, rebuild the TUI from a clean release worktree, inspect MCP/addon module sources and tool surface, and report evidence safely with secrets redacted.

Nested lingtai-dev-guide reference for turning one-off experience into durable skills: when to write a skill vs leave it as notes, the router-vs-nested- reference pattern, distilling experience (归一) and session journals into triggered/executable SKILL.md files, de-privatizing and parameterizing local paths and human-specific details, a lightweight pre-publish benchmark/checklist, grooming shared-library candidates, and PR-ready skill cleanup.

Nested lingtai-dev-guide reference for contribution workflow: issue/worktree/PR discipline, stale-worktree cleanup, daemon decomposition, portfolio sweeps, repo-specific build/test commands, skill changes, and anatomy maintenance.

Operational guide for the `mcp` capability — register, activate, update, deregister, and troubleshoot MCP (Model Context Protocol) servers in your agent. Single source of truth for both generic MCP setup AND the five kernel-curated LingTai addon MCPs (`imap`, `telegram`, `feishu`, `wechat`, `whatsapp`, `cloud_mail`). Reach for this manual when: - The human asks to install, set up, configure, or remove an MCP server. Decision tree: is it kernel-curated (imap/telegram/feishu/wechat/whatsapp/cloud_mail) → the `addons:` + `init.json mcp.<name>` workflow using modules shipped inside the `lingtai` wheel is here. Is it third-party → the registry route OR the legacy `mcp/servers.json` route, both documented here. - The human asks to set up the `imap` / `telegram` / `feishu` / `wechat` addon, or any LingTai email/chat integration. **Step 1 is always**: read the curated-addon setup section and, when exact config fields are needed, inspect the shipped module docs/resources or the catalog

**Read this before running long-lived agent/coding CLIs (`claude -p`, `codex exec`, `opencode run`, Cursor agent CLI), or before setting up cron, launchd, systemd timers, crontab jobs, or scheduled reminders.** Router for Bash-related operational depth beyond the bash tool schema: async + poll discipline for long-running child agents, host-scheduler setup, LingTai wake-by-mailbox-drop, script hygiene, one-shot `.notification/cron.json` reminders, debugging silent jobs, and safe cleanup. Start here for any long-running agent CLI, time-driven recurring work ("every hour", "weekdays at 9", "remind me later"), or when a scheduled job misbehaves.

Nested daemon-manual reference for daemon API details and CLI backends: daemon(action=list), claude/claude-p/codex/opencode behavior, backend_options flag passing, preset/capability inheritance, and Codex modal capabilities.

Nested daemon-manual reference for polling cadence, stall heuristics, anti- patterns, backend-specific polling notes, and setting reminders before resting while daemon work remains pending.

Operational router for the `daemon` tool: inspect slow/stuck/failed emanations, read daemon artifact folders, choose polling cadence, avoid reclaiming on a hunch, understand `daemon(action="list")`, use CLI backends and `backend_options`, and clean up daemon footprint. Read this after dispatching daemon work that is slow, failed, timed out, or needs backend-specific reasoning.

Nested system-manual reference for expanded LingTai procedure/action guidance. Read via the `system-manual` router when resident procedures are too compact and you need details about progressive disclosure, responsiveness, external side-effect authorization, choosing bash vs daemon vs avatar vs MCP, depositing work into pad/knowledge/skills/character, idle/lifecycle procedure, molt checklist, skill routing, web/file/media/artifact handling, standalone HTML deliverables, sharing artifacts, issue reporting, and resident procedures maintenance. This is a nested skill-reference under `system-manual`, not a standalone catalog skill; its folder may carry scripts/assets as procedure guidance grows.

Second-layer router for LingTai's progressive-disclosure operating manuals. Read this when resident substrate/procedures are too compact and you need to route to the right lower reference. References include `reference/substrate-manual/SKILL.md` for the expanded substrate/body/lifecycle/ communication/memory/idle/system-operations model; `reference/procedures-manual/SKILL.md` for the expanded procedures/action discipline, skill routing, responsiveness, deliverables, artifact sharing, and issue reporting guidance; and `reference/sqlite-log-query/SKILL.md` for SQLite/log.sqlite runtime trace inspection and trajectory/anomaly mining from event traces. Also route here for lifecycle operations, notification handling, molt/memory questions, MCP/addon ownership, preset tiers, collaboration/network topology, resident prompt design, and the `system` tool actions.

Operational guide for the `email` tool — LingTai email protocol within your `.lingtai/` network. Covers send/check/read/dismiss/reply/reply_all/ search/delete/archive/contacts, reply discipline (always reply on the channel the message arrived on), addressing (bare paths like `human`, `mimo-1`), self-send for persistent notes that survive molt, time capsules (delayed self-send via `delay`), scheduled email (recurring alarms via `schedule={action: "create", ...}`), the unread digest notification contract, and addon ownership. This is for INTERNAL email only — for real internet email via IMAP/SMTP, see the `mcp-manual` skill (the lingtai-imap addon owns that surface).