Obsidian Mind is an Obsidian vault that stores notes, links, indexes, and other accumulated context so AI coding agents can retain memory across sessions. It is designed for Claude Code and also supports Codex CLI and Gemini CLI through shared hooks, commands, and vault data. Its catalogue add-ons package the vault’s commands, agents, skills, hooks, instructions, settings, plugin, and MCP integration.
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.
npx agentmods add commands/breferrari/obsidian-mind/om-correctgit clone --depth 1 https://github.com/breferrari/obsidian-mindWrote 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/breferrari/obsidian-mind/om-correct)<a href="https://agentmods.dev/commands/breferrari/obsidian-mind/om-correct"><img src="https://agentmods.dev/badge/commands/breferrari/obsidian-mind/om-correct.svg" alt="Measured on agentmods" 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 | $0.00043 | $0.01290 |
| Opus 5 | $0.00022 | $0.00645 |
| Sonnet 5 | $0.00009 | $0.00258 |
| Haiku 4.5 | $0.00004 | $0.00129 |
Grade A, and why
om-correct 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 5d 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.
How it starts
The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/om-correct
/om-correct "<the corrected fact, stated as it is now true>"
The acting half of Write-Correctness Law 2. Laws 1 and 2 already say a corrected fact must be swept out of every note restating it; this is the thing that does it.
Run it at the moment of correction, not on a schedule. /om-vault-audit is periodic, whole-vault and structural, and takes no argument. A sweep needs the corrected fact as input, because nothing in the vault can resolve two contradicting notes — nothing in it says which is true. Only the person making the correction knows, which is why the fact is an argument and not a discovery.
Hard rails
- A historical note is never edited. Not carefully, not with a callout. It is reported and preserved. See the asymmetry below.
- Never delete. Restatements are replaced with a link to the source, not removed.
- Never commit. Leave the working tree for review.
- No authoritative note, no edits. If the single source cannot be identified, the pass reports and stops rather than guessing.
1. Find
Both arms, always. Neither alone is sufficient.
- Semantic —
qmd query "<the corrected fact>"(index name fromvault-manifest.json). This is the arm that finds the paraphrase: the same claim in someone else's words, with none of the same strings. That half is invisible to grep and indistinguishable from correct prose to every structural check, and it is the half most likely to survive a manual sweep, because the sweeper stops when grep goes quiet. - Literal — grep the distinctive strings from the fact: version numbers, tool names, counts, identifiers. This is the arm that finds a bare
v8.3.2that means nothing to an embedding.
2. Classify
This is the design. The search is the easy part.
| class | what it is | what happens |
|---|---|---|
| AUTHORITATIVE | the single-source note Law 1 designates | the correction is applied here, and only here |
| RESTATEMENT | a living note asserting the fact, which per Law 1 should be linking to the source instead | the assertion is replaced with a link — fixes today's stain and prevents the next |
| HISTORICAL | a note that correctly records what was believed at the time | preserved, and reported so the omission is visible |
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.
- 5d ago First seen · 88 lines · 43 tokens per session scan A 8f081b52dbb7
om-correct is a command published in the GitHub repository breferrari/obsidian-mind (4,599 stars, last pushed 2d ago), licensed MIT. It adds 43 tokens to every session and 1,290 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
kb-archive
Archive, detach, purge, or rename KB objects while keeping registry, index, and links consistent.
kb-init
Initialize or rebuild a vault-first, project-scoped Obsidian KB under Research/{project-slug}/ for the current repository.
kb-sync
Run deterministic KB maintenance to refresh scaffold integrity, registry, index, daily note, and runtime binding summary.
kb-status
Summarize the current bound project KB status, including registry counts and key project note paths.
kb-ingest
Ingest external material into Sources/ inside the bound project KB, then update registry, index, and daily note as needed.
kb-promote
Promote durable content from Daily or source notes into canonical Knowledge, Experiments, Results, Results/Reports, or Writing notes.