obsidian-second-brain turns an Obsidian vault into persistent, searchable memory for Claude Code and other command-line coding agents, storing knowledge as linked Markdown notes. It is for developers, founders, writers, and researchers who want agents to retain project context across sessions. Its catalogue entries provide commands, hooks, a plugin, a skill, and instructions for capturing, finding, and maintaining that memory.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/eugeniughelbur/obsidian-second-brainWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/eugeniughelbur/obsidian-second-brain/obsidian-catchup)<a href="https://agentmods.dev/commands/eugeniughelbur/obsidian-second-brain/obsidian-catchup"><img src="https://agentmods.dev/badge/commands/eugeniughelbur/obsidian-second-brain/obsidian-catchup/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/eugeniughelbur/obsidian-second-brain/obsidian-catchup"><img src="https://agentmods.dev/badge/commands/eugeniughelbur/obsidian-second-brain/obsidian-catchup.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00048 | $0.01173 |
| Opus 5 | $0.00024 | $0.00587 |
| Sonnet 5 | $0.00010 | $0.00235 |
| Haiku 4.5 | $0.00005 | $0.00117 |
Grade A, and why
obsidian-catchup scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Use the obsidian-second-brain skill. Execute /obsidian-catchup $ARGUMENTS:
The optional argument is a timeframe filter: today, week, or all (default: all unprocessed).
This is the laptop-side companion to the Telegram journal bot (integrations/telegram-journal/). The bot captures on the go (voice / text / image / PDF / link) and appends each one to a catchup.md queue in the vault. This command is where those captures become real, integrated vault knowledge - on your schedule, with you driving. It is a PULL: nothing was processed automatically, so nothing surprises you.
- Read
_CLAUDE.mdfirst if it exists in the vault root (for vault paths + conventions). - Read
catchup.mdin the vault root - the queue the bot fills. Each line is- [ ] date time | kind | summaryplus, only when the bot saved the item somewhere, a trailing| -> wherelink. An unchecked- [ ]is unprocessed;- [x]is already done. - Collect the unchecked items. Apply the timeframe filter if given (
today= today's date;week= the last 7 days;all= every unchecked). If there are none, say "nothing to catch up on" and stop. - Show the user a tight list, grouped by age (Today / This week / Older): time, kind, summary, and the
[[link]]to where it landed (when present). Flag stale ones explicitly (e.g. "captured 9 days ago - still relevant?"), since a capture that mattered on a walk may be dead now. - Process the items WITH the user (this is a together-review, never an autonomous sweep). For each item - or a sensible batch - open the linked note/entry (or work from the queue summary when there is no link), read the actual captured content, and propose one of:
- Integrate - fold it into the right existing or new note (person / project / concept / decision), following
references/ai-first-rules.md: update existing notes, fill and link, reconcile contradictions - exactly as/obsidian-savewould. Prefer updating an existing note over creating a near-duplicate. - Keep as-is - it is already filed fine (e.g. a daily-note thought); just acknowledge it.
- Discard - stale or junk; remove the stray entry from where it landed. Surface your proposal and let the user confirm or redirect before you write anything.
- Integrate - fold it into the right existing or new note (person / project / concept / decision), following
- After an item is handled, check it off in
catchup.md: change its- [ ]to- [x]and append(processed YYYY-MM-DD). Never delete queue history - it is the record of what came in. - End with a one-line summary: N integrated, M discarded, K left unprocessed.
The split this enforces: the phone is for fast, dumb capture; the laptop is where you think and integrate. Same brain, two speeds - not two silos.
AI-first rule: Every note created or updated by this command MUST follow references/ai-first-rules.md - ## For future agent preamble, rich frontmatter (type, date, tags, ai-first: true, plus type-specific fields), recency markers per external claim, mandatory [[wikilinks]] for every person/project/concept referenced, sources preserved verbatim with URLs inline, and confidence levels where applicable. If that path does not resolve from your working directory, search upward for it; if you still cannot read it, say so before writing rather than producing a note that silently skips the rule. The vault is for future agent retrieval - not human reading.
Anti-fabrication: Search exhaustively before claiming any note, person, or file is absent - false absence is the most common failure mode - and never invent facts, entities, or dates (mark unknowns as TBD). See the anti-fabrication and search-completeness hard rules in references/ai-first-rules.md.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 35 lines · 48 tokens per session scan A 8ce291ac8b0d
obsidian-catchup is a command published in the GitHub repository eugeniughelbur/obsidian-second-brain (4,376 stars, last pushed 3d ago), licensed MIT. It adds 48 tokens to every session and 1,173 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
patterns
Scan recent sessions for recurring patterns and turn them into captures (rules, concept notes, writing seeds).
second-brain-mapping
Map your vault: extract structured metadata from every typed file, surface cross-doc insights, optionally build a knowledge graph.
setup-vault-types
Configure which document types your vault uses (journals, books, meetings, clients, etc.) and scaffold extractors.
cierre
A sales call just ended: turn its transcript into the full follow-up (CRM, tasks, email draft, reminder, coaching).
daily-journal
Daily journal interview and entry creator with emotional floor tagging.
deconstruct
First-principles analyst: surface hidden assumptions, find foundational truths, rebuild from scratch.