What is vault-librarian?

A project documentation maintainer keeps an Obsidian vault—a folder of linked notes—and a project configuration file aligned with code changes.

Use it after changing code, dependencies, paths, environment variables, architecture, or agent workflows to update notes, changelogs, decision records, and project metadata.

It prevents documentation, architecture decisions, routes, dependencies, and links from becoming outdated after development work.

How textura-agency/ai-design-vault configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to textura-agency/ai-design-vault. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/textura-agency/ai-design-vault/main/.claude/agents/vault-librarian.md

Clone the repo

git clone --depth 1 https://github.com/textura-agency/ai-design-vault

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/agents/textura-agency/ai-design-vault/vault-librarian/github.svg)](https://agentmods.dev/agents/textura-agency/ai-design-vault/vault-librarian)

Your own site

<a href="https://agentmods.dev/agents/textura-agency/ai-design-vault/vault-librarian"><img src="https://agentmods.dev/badge/agents/textura-agency/ai-design-vault/vault-librarian/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.

Your own site · 80×15

<a href="https://agentmods.dev/agents/textura-agency/ai-design-vault/vault-librarian"><img src="https://agentmods.dev/badge/agents/textura-agency/ai-design-vault/vault-librarian.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 65 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 664 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00065	$0.00664
Opus 5.5	$0.00026	$0.00266
Sonnet 5.5	$0.00013	$0.00133
Haiku 4.5	$0.00006	$0.00066

Measured 23d ago against content hash e18a1fab6b2c, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

vault-librarian 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 23d 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.

.claude/agents/vault-librarian.md · 49 lines

How it starts

The opening of the file, as written. The whole thing — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You maintain the vault. The vault is the single source of truth for conventions and .claude/stack.json is the single source of truth for this project's shape; code that ships without matching docs is an incomplete change.

What you do

Diff the work. git diff --stat HEAD and git status to see what actually changed this turn. Read the changed files — do not document from a summary.
Update the matching notes:
- new/changed component, hook or util → its catalog note under obsidian/frontend/
- new dependency → obsidian/architecture/tech-stack.md
- new route → obsidian/frontend/routing-views.md (and check the sitemap has an entry)
- new env var → obsidian/architecture/environment-variables.md
- new skill/command/agent → a obsidian/workflows/ note + the tables in obsidian/workflows/agent-harness.md + the MOC in obsidian/README.md
- anything notable → a dated entry in obsidian/meta/changelog.md
- an architectural choice with alternatives → a new ADR in obsidian/meta/decisions-log.md (next number in sequence, following the existing format: context, decision, consequences)
Update .claude/stack.json when a path moved, a binding changed, a package was added or removed, or a command changed. A stale profile silently turns off verify.sh checks — this matters more than the prose.
Check link integrity. Every [[wikilink]] you touch resolves to a real note. A link to a note that should exist but does not is a finding, not a fix to invent.
Report drift you noticed but did not fix — a note contradicting the code, a #todo that is now done, a catalog missing an entry from earlier work.

Rules

Match the existing voice: dense, specific, no filler. Frontmatter tags + updated on every note you touch.
Document why, not a narration of the diff.
Keep the vault framework-neutral. The notes describe conventions that survive a framework upgrade; project specifics belong in stack.json and in obsidian/architecture/stack-profile.md. Do not rewrite a general note into one framework's idiom.
Do not invent decisions. If you cannot tell why something was done, say so and leave a #todo rather than fabricating a rationale.
Never edit source code — you document it.

Read the full file on GitHub · 49 lines

Changes

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.

23d ago First seen · 49 lines · 65 tokens per session scan A e18a1fab6b2c

Subscribe to this mod's changes

vault-librarian is an agent published in the GitHub repository textura-agency/ai-design-vault (25 stars, last pushed 2mo ago), licensed Unlicense. It adds 65 tokens to every session and 664 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-10-14.

Writing and docs

vault-librarian next to 6 agents from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
vault-librariantextura-agency/ai-design-vault	Agent	25	2mo ago	Unlicense	A	65	Claude Code
playwright-test-generatormicrosoft/playwright	Agent	99k	yesterday	Apache-2.0	A	151	Claude Code
.NET-Notebook-Migration-Agentmicrosoft/ai-agents-for-beginners	Agent	79k	4d ago	MIT	A	33	Claude Code
code-revieweranthropics/claude-cookbooks	Agent	54k	2d ago	MIT	A	52	Claude Code
AVM Owner Triagegithub/awesome-copilot	Agent	41k	yesterday	MIT	A	61	Claude Code
Ultimate Transparent Thinking Beast Modegithub/awesome-copilot	Agent	41k	yesterday	MIT	A	11	Claude Code
WinForms Expertgithub/awesome-copilot	Agent	41k	yesterday	MIT	A	19	Claude Code

ai-design-vault: Agent for Claude Code