agent-second-brain

Agent Second Brain An always-on second brain you talk to. Voice note in Telegram → typed, linked knowledge in your Obsidian vault. Runs 24/7 on a $5 VPS and the Claude subscription you already pay for — zero per-token API bills. 🇷🇺 Русский • Полная инструкция • VPS guide • autograph engine --- [!IMPORTANT] **Since June 15, 2026, headless `claude -p` runs bill against a separate paid Agent SDK credit.** v3.0 sidesteps that completely: it drives one long-lived *interactive* Claude Code session — the same thing you run in a terminal, used exactly the way the subscription is meant to be used. No headless calls in the hot path (enforced by a CI guard), no per-request billing, no surprise invoice. Start here You want to… | Go to | ---|---| Understand what this thing is | Why I built this | Install it on a fresh VPS in one command | Quick start | Upgrade an existing v1 / v2 install | Upgrading | Пошаговая инструкция на русском, для новичков | docs/setup-guide.ru.md | Just want the memory engine for your own vault | autograph → | See how the persistent-session trick works | How it works | Why I built this Every productivity system dies the same way. You set it up on a Sunday, use it for two weeks, then slowly stop — because maintaining the system is more work than the work itself. Voice memos you never re-listen to. Ideas that drown in chat history. A vault of markdown files that nobody, including you, can navigate a month later. The fix isn't another app. It's removing the organizing step entirely: **you talk, the agent files**. And it has to be *yours* — I wasn't going to pipe my private notes, clients, and goals through somebody's SaaS. Everything here runs on your own server, lands in your own Obsidian vault as plain markdown, and is small enough to actually read before you trust it with your life. What talking to it looks like No commands to memorize, no categories to pick, no app to open. Telegram is the whole interface. Philosophy **Voice-first.** Capture has to be cheaper than forgetting, or the system dies. A voice note costs five seconds. **The vault is the source of truth.** Everything lives as plain markdown in *your* Obsidian vault on *your* server. Delete the agent tomorrow — you keep everything, readable forever. No lock-in, no export button needed. **Memory that forgets, like yours.** Storage is not memory. Knowledge decays on the Ebbinghaus curve, fades through five tiers, and resurfaces when it matters — so the graph stays sharp instead of becoming a landfill. **Interactive session, by the rules.** One persistent Claude Code session in tmux, driven the way a human drives it. No headless `claude -p` anywhere in the hot path — a CI guard fails the build if anyone tries. **Small enough to read.** One Python process, a handful of modules, 220+ tests. You can audit the thing that reads your private notes in an evening. What it does 🎙 Total captureVoice (Deepgram, seconds), text, photos, documents, videos, forwarded posts, whole albums — the agent reads files itself and files the takeaways. Nothing you send is ever silently dropped. 🧠 Knowledge graph memoryPowered by autograph: typed cards, wiki-links, Ebbinghaus decay across five tiers, automatic MOC indexes, health scoring, link repair, dedup. The most thorough open-source memory layer you can drop on a vault. ⏰ Self-managed routines"Remind me Friday at 3pm", "every weekday at 18:30 check my inbox folder" — the agent schedules its own cron jobs from plain language. One-shots, intervals, full cron expressions. No external task manager. 🌙 Nightly processingAt 21:00 your time it classifies the day's entries, writes vault cards, updates goals and long-term memory, rebuilds the graph — and sends you a daily report. 🔌 Claude Code, but for PKMIt IS Claude Code under the hood — so drop any MCP server into mcp-config.json, add any skill into vault/.claude/skills/, and your second brain grows new abilities. Like Claude Code through Telegram — for knowledge, not code. 🩺 Self-healingA watchdog recovers a wedged session, a daily doctor sends a 🟢/🔴 canary report, broken jobs disable themselves and tell you. It just stays up. The memory engine: autograph The part that makes this a *brain* rather than a logger is **autograph** — a typed memory layer for always-on agents, shipped here as a skill and also usable standalone on any Obsidian vault. - **Typed graph** — every card carries a type (note, contact, project, CRM), a description for retrieval, tags, status; one `schema.json` rules them all - **Ebbinghaus decay** — strength `1 + ln(access_count)`: each touch slows forgetting; contacts fade in ~100 days, dailies in ~25 - **Five tiers** — core → active → warm → cold → archive; touching a card promotes it back up - **Random recall** — occasionally an archived card resurfaces next to something current. Sometimes noise. Sometimes the best idea you forgot you had. - **Self-maintenance** — orphan detection, broken-link repair, dedup into `.trash/`, MOC generation, a 100-point health score. You never run vault chores again. Tier | What happens | ------|-------------| **Core** | Always in context: current projects, active clients, key goals | **Active** | Checked regularly: recent ideas, ongoing threads | **Warm** | Found when you search | **Cold** | Surfaces only in deep searches | **Archive** | Almost gone — but randomly recalled for creative collisions | How it works The bot never spawns `claude` per message. It keeps **one long-lived interactive Claude Code session** alive in a tmux pane and types prompts into it — which is why a 24/7 agent runs on a flat Pro/Max subscription instead of metered API tokens. Scheduled jobs fire in a second, isolated session, so a reminder going off never interrupts your conversation. A cross-process lock serializes everything; a watchdog restarts whatever wedges. Privacy is explicit: **voice audio** goes to Deepgram for transcription, **text** goes to Anthropic through your subscription, **everything else stays on your server**. The vault never leaves the machine. What it costs Service | Cost | ---------|------| Claude Pro | $20/mo | VPS (any cheap one) | ~$5/mo | Deepgram | free tier ($200 credit) | **Total** | **~$25/mo, flat** |