本文へ移動
cccskills
無料GitHub で公開

health-sync

Pull fresh messages from a set of health-related Telegram chats (medicine, supplements, longevity, fasting) into an Obsidian vault. Incremental and idempotent by message id, and the manual counterpart of the weekly job. Triggers: "/health-sync", "pull the health chats".

tonydzi/second-brain-starter-kit
インストール方法を見る

含まれるファイル（1）

  • SKILL.md5.6 KB

SKILL.md（原文）

インストールする前に、エージェントに与えられる指示の中身を確認できます。

OBJECTIVE: Pull fresh messages from Anton's 7 Telegram health chats and incrementally fold them into the Obsidian vault (day-ledgers + atomic notes + MOC + dashboard), on demand. Fully idempotent (dedup by Telegram message id). Identical pipeline to the weekly task health-weekly-sync; the only difference is this fires when Anton asks.

CONTEXT:

  • 7 source chats (registry: $IMPORTS_ROOT/health/chats.json). ALWAYS pull with the Telegram MCP using account="default" (@work_acct_a); the other account errors (GEN-ERR-862) on these basic groups.
    • <YOUR_CHAT_ID> medicine-health-main
    • <YOUR_CHAT_ID> blood-pressure
    • <YOUR_CHAT_ID> longevity-weightloss-fasting
    • <YOUR_CHAT_ID> vitamins-bads
    • <YOUR_CHAT_ID> healing-head
    • <YOUR_CHAT_ID> medicine-for-mind
    • <YOUR_CHAT_ID> fasting-community
  • Vault home: $OBSIDIAN_VAULT/01-Conversations/Telegram/Health/ (per-chat folders + posts\ + _Health-MOC.md). Dashboard: _Dashboards\Health-Dashboard.html. Scripts + watermarks + raw: $IMPORTS_ROOT/health/.
  • Watermark per chat (last imported msg id) = $IMPORTS_ROOT/health/watermarks.json (keyed by chat_id as string).
  • Provenance: health data is NOT private (Anton's explicit instruction) — do NOT tag #private. Anton's own messages -> origin: anton (+ #anton-original); forwarded -> origin: external (+ forwarded_from); community members (fasting-community) -> origin: external.

STEPS:

  1. Health-check the Telegram MCP: call mcp__telegram__get_me. If it errors, STOP and report "Telegram MCP is down — skipped".
  2. Read watermarks: load $IMPORTS_ROOT/health/watermarks.json -> {chat_id: max_msg_id}.
  3. For EACH of the 7 chats: call mcp__telegram__get_new_messages_since(chat_id=<id>, min_id=<watermark for that id>, account="default") (or get_history limit 100 then keep id > watermark). Collect the message objects with id > watermark.
  4. For each chat, write its new messages (id > watermark) VERBATIM to $IMPORTS_ROOT/health/raw/_new/<slug>.json as a JSON array of the raw message objects (every field: id, sender, sender_id, date, out, text, media, forwarded, reply_to, grouped_id, pinned, edited, action). NEVER truncate or paraphrase text. If a chat has no new messages, write [] (or skip its file).
  5. Run: python $IMPORTS_ROOT/health/merge_health_raw.py (merges raw_new* into raw*, dedup by id; writes last_merge.json).
  6. Read $IMPORTS_ROOT/health/last_merge.json. If total_new == 0: STOP and report "No new health messages." Do nothing else.
  7. BACKUP FIRST (vault-backup-rule): run python $IMPORTS_ROOT/vault_backup.py. If it reports a git index.lock from a concurrent process, note it but proceed — the import is purely ADDITIVE to the Health folder (reversible by deleting new files); never force-remove another process's lock.
  8. Run: python $IMPORTS_ROOT/health/run_health_pipeline.py (parse -> generate -> moc -> dashboard -> validate; expects BROKEN=0).
  9. Robocopy staging into the vault (PowerShell): robocopy "$IMPORTS_ROOT/health/staging" "$OBSIDIAN_VAULT/01-Conversations/Telegram/Health" /E /XO /NFL /NDL /NJH /NJS (exit code 0-7 = success.)
  10. Reindex for RAG (so the new notes are searchable): python $IMPORTS_ROOT/brain_embed_update.py --wait-gpu 10 (or rely on the nightly Brain Reindex at 04:00 if the GPU is busy).
  11. Report: new messages per chat, new atomic notes / ledgers created, the new watermarks (max ids), and confirm BROKEN=0. End with a 🧒 In plain words recap (messages TO Anton only).

CONSTRAINTS:

  • WINDOWS cp1252: never print Cyrillic to python stdout (crashes) — write results to UTF-8 files and Read them; keep python stdout ASCII (counts/slugs only).
  • INCREMENTAL ONLY by design, but the pipeline is a full deterministic rebuild of staging from the archive (idempotent) + robocopy /XO — it never duplicates and never deletes existing vault notes.
  • Raw files may have a UTF-8 BOM (the scripts read utf-8-sig) — fine.
  • Faithful raw import: write message text VERBATIM into raw_new files. The atomic-note bodies are the raw text; do not summarize.

RELATION (do not duplicate):

  • Scheduled twin: $USERPROFILE/.claude/scheduled-tasks/health-weekly-sync/ (Mondays) — same scripts.
  • First-time backfill design + counts: memory [[health-import]].
  • Sibling pattern: [[faaa-sync]] (CRM calls). Reuses the FAAA weekly-sync architecture.
<!--kit-footer-->

Like this skill? It is one of 100 in second-brain-starter-kit: the second brain we built for ourselves and run every day at Palo Alto AI Research Lab. Install the whole set with npx skills add tonydzi/second-brain-starter-kit. Everything is open source and free, so take what you need.

Flagships worth a look on their own: secondop-panel (a second opinion from a panel of external models), claude-memory-tidy (stop your agent's memory from rotting), telegram-mcp-kit (your own Telegram over MCP in about 15 minutes).

Author: Anton Dziatkovskii, Palo Alto AI Research Lab. Telegram @tonydzi - WhatsApp +1 341 222 9178 - X @Tony_Stef_

Engineers: want to test-drive this setup? Message me. I hand out free starter seeds to engineers who test and report back, and custom skill requests are welcome.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

ログインしてレビューを書く

同じリポジトリのスキル

概要と使いどころ

03

無料

Run an autonomous consensus round between agent peers on several machines: they negotiate a decision over a shared channel and execute it without a human relaying messages. Use when multiple machines must agree on a change. Triggers: "/03", "work it out among yourselves", "find consensus". Wakes the human only for irreversible actions or a deadlock.

日本語の概要は準備中です。原文の説明を表示しています。

tonydzi/second-brain-starter-kit82026年10月12日 更新

1

無料

Recover a session after a hard crash (app died, machine rebooted, context lost mid-work): replay the turn-state ledger for the last requests, files and decisions, run a green/red health ping over the system map, file sync and MCP connectors, then rebuild the previous session history for pickup. Triggers: "/1", "resume after crash", "wake".

日本語の概要は準備中です。原文の説明を表示しています。

tonydzi/second-brain-starter-kit82026年10月12日 更新

agenda

無料

Show the operator's day: today's calendar plus what needs them personally (leads due, people awaiting a reply, deadlines), pulled from Calendar and Telegram and ranked. Read-only: sends and commits nothing. Triggers: "/agenda", "what's on today", "my agenda".

日本語の概要は準備中です。原文の説明を表示しています。

tonydzi/second-brain-starter-kit82026年10月12日 更新

ai-slop

無料

Rewrite machine-written text so it reads like a busy human wrote it: cut 50-70%, break the even rhythm, drop AI-tell words and filler, keep facts and protected terms. Use before publishing any post, reply or email drafted by a model. Triggers: /ai-slop, /slop, humanize, anti-slop, make it shorter and alive.

日本語の概要は準備中です。原文の説明を表示しています。

tonydzi/second-brain-starter-kit82026年10月12日 更新

alfa-search-recall-deepresearch

無料

Decision protocol for strategic work: recall the knowledge vault, run a gap analysis, emit a deep-research prompt for external tools, then synthesize the returned reports into a decision memo. Use before any new product, feature, market, pricing, GTM or architecture call, where local recall alone is not enough. Triggers: "/alfa-search-recall-deepresearch", "alpha protocol", "R+DR".

日本語の概要は準備中です。原文の説明を表示しています。

tonydzi/second-brain-starter-kit82026年10月12日 更新

alpha-credit

無料

- КРЕДИТ АВТОРУ ИДЕИ: совет из комментария (TG @ClawRus / @ClawInga, комменты под постами Антона в FB, треды и issue на… Триггеры: “/alpha-credit“, “/credit“, “кто нам это посоветовал“, “запиши совет“, “поблагодари советчиков“, “кому мы должны спасибо“, “кредит автору“, “чей это был совет“, “credit the author“

日本語の概要は準備中です。原文の説明を表示しています。

tonydzi/second-brain-starter-kit82026年10月12日 更新

tonydzi のスキルをすべて見る

このスキルの問題を報告する