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
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/eugeniughelbur/obsidian-second-brainnpx agentmods add commands/eugeniughelbur/obsidian-second-brain/obsidian-reindexWrote 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-reindex)<a href="https://agentmods.dev/commands/eugeniughelbur/obsidian-second-brain/obsidian-reindex"><img src="https://agentmods.dev/badge/commands/eugeniughelbur/obsidian-second-brain/obsidian-reindex/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-reindex"><img src="https://agentmods.dev/badge/commands/eugeniughelbur/obsidian-second-brain/obsidian-reindex.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 | $0.00013 | $0.00720 |
| Opus 5 | $0.00006 | $0.00360 |
| Sonnet 5 | $0.00003 | $0.00144 |
| Haiku 4.5 | $0.00001 | $0.00072 |
Grade A, and why
obsidian-reindex 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 12d 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-reindex:
The semantic index is incremental. An existing, current index is cheap to refresh because only new and changed notes are embedded again.
-
Read
_CLAUDE.mdfirst to find the vault path. Resolve it to an absolute path and substitute it forVAULT_PATHbelow. The skill root was given at session start as Skill root; substitute its absolute path forSKILL_ROOT. -
Measure coverage before changing the index:
uv run --directory "SKILL_ROOT" python -c 'import json, sys; from pathlib import Path; sys.path.insert(0, str(Path(sys.argv[1]) / "integrations" / "obsidian-mcp-server")); from vault_ops import index_coverage; print(json.dumps(index_coverage(Path(sys.argv[2]).expanduser().resolve())))' "SKILL_ROOT" "VAULT_PATH"Report
indexed,scanned,missing, andpct_missing. Ifindexis false, say that no readable semantic index exists yet instead of presenting0/0as coverage. -
Tell the user that the build is incremental, then run it from the skill root:
uv run --directory "SKILL_ROOT" python scripts/eval/semantic_search.py --path "VAULT_PATH" --buildPreserve the command's exit status and capture its stderr. If it exits nonzero, stop and show the actionable backend error. Do not claim the index was refreshed and do not report an after-state as success. The common failure is Ollama not running or the configured embedding model not being pulled; use the runtime and model named by the command's own output rather than assuming the defaults.
-
On success, run the coverage command from step 2 again.
-
Report the result:
- Coverage before and after as
indexed/scanned, with missing count and percentage - The builder's
newcount from its[semantic] indexed ...summary as notes newly embedded or refreshed - Cached, excluded, degraded, and dropped counts when nonzero
- Every degraded or dropped path printed by the builder, because those notes are not fully searchable by meaning
- Coverage before and after as
If coverage did not improve, say so plainly and use the builder output to distinguish a current index from failed or excluded notes. This command updates only .obsidian-semantic-index.json; it does not modify Markdown notes.
Anti-fabrication: Report coverage and build counts exactly as the tools emit them. Never infer a successful refresh from an unchanged file or a zero exit status alone when the builder reports dropped notes.
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.
- 12d ago First seen · 41 lines · 13 tokens per session scan A dbc88b10d234
obsidian-reindex is a command published in the GitHub repository eugeniughelbur/obsidian-second-brain (4,392 stars, last pushed 4d ago), licensed MIT. It adds 13 tokens to every session and 720 once invoked, about $0.0001 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
cierre
A sales call just ended: turn its transcript into the full follow-up (CRM, tasks, email draft, reminder, coaching).
deconstruct
First-principles analyst: surface hidden assumptions, find foundational truths, rebuild from scratch.
instinct-import
Import a portable YAML instinct pack with confidence-gated merge (Instinct Engine).
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.