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 instructions/uninhibited-scholar/obsidian-loop-wiki/agents-mdgit clone --depth 1 https://github.com/uninhibited-scholar/obsidian-loop-wikiWrote 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/instructions/uninhibited-scholar/obsidian-loop-wiki/agents-md)<a href="https://agentmods.dev/instructions/uninhibited-scholar/obsidian-loop-wiki/agents-md"><img src="https://agentmods.dev/badge/instructions/uninhibited-scholar/obsidian-loop-wiki/agents-md.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 | $0.00768 | $0.00768 |
| Opus 5 | $0.00384 | $0.00384 |
| Sonnet 5 | $0.00154 | $0.00154 |
| Haiku 4.5 | $0.00077 | $0.00077 |
Grade A, and why
obsidian-loop-wiki AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — obsidian-loop-wiki
Drop this file in your project root so Codex CLI, Gemini CLI, or any agent that reads an instructions file follows the same loop as the Claude Code skill. (Claude Code users: run
./install.shinstead — it installs the richer skill version.)
You maintain an Obsidian knowledge base as a verifiable loop. Raw material goes in; you compile it into typed, cross-linked nodes that form a graph, and every run leaves an audit trail. The point of the discipline below is that an agent told to "just organize my notes" produces AI-generated clutter — drifting names, orphan pages, unsourced claims. The loop prevents that.
Locate the vault first
The vault path is in the WIKI_PATH environment variable. Never guess it.
echo "$WIKI_PATH" && ls "$WIKI_PATH"
If WIKI_PATH is unset, ask the user for their vault path. If the path has no
SCHEMA.md, it's uninitialized — scaffold it from templates/ before doing
anything. Otherwise read SCHEMA.md, index.md, log.md first; SCHEMA.md's
rules override your instincts.
Node types (keep separated)
raw/— sources. Append-only, read-only. Never edit originals.concepts/— durable ideas recurring across sources.entities/— tools, people, companies, projects.comparisons/— "X vs Y" pages that accrete over time.queries/— answers worth keeping.moc/— Maps of Content (reading routes, not listings).drafts/— outputs.
Ingest an article / organize the inbox
- Read the source; extract core themes. Do not modify originals.
- Create/update the right typed nodes per
SCHEMA.md. - Cross-link with
[[wikilinks]]— every new node links to ≥1 existing node. An orphan node is a failed run. This single rule is what makes a graph exist instead of a pile of islands. - Cite sources:
(source: raw/articles/<file>.md). - Flag inferences/unconfirmed claims as
待验证(to-verify). Never launder a guess as fact. - Update
index.md(notable nodes) and the relevantmoc/map. - Append a dated
log.mdentry: source, files created/updated, to-verify items. - Run the verification pass below and report exactly what you created/updated. If unsure where something belongs, stop and ask — a wrong classification pollutes the graph silently.
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.
- 4d ago First seen · 68 lines · 768 tokens per session scan A 797ed0e87428
obsidian-loop-wiki AGENTS.md is an instructions file published in the GitHub repository uninhibited-scholar/obsidian-loop-wiki (2 stars, last pushed 2mo ago), licensed MIT. It adds 768 tokens to every session, about $0.0038 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 2026-08-31.
Other instructions, from other repositories
obsidian-mind AGENTS.md
AGENTS.md instructions for breferrari/obsidian-mind, covering obsidian mind, hooks, commands, memory and reaching this vault from another repo.
obsidian-mind GEMINI.md
Gemini CLI instructions for breferrari/obsidian-mind, covering obsidian mind, hooks, commands, memory and reaching this vault from another repo.
obsidian-mind CLAUDE.md
Claude Code instructions for breferrari/obsidian-mind, covering obsidian mind, skills & capabilities, custom slash commands, vault structure and obsidian cli.
obsidian-llm-wiki CLAUDE.md
Claude Code instructions for green-dalii/obsidian-llm-wiki, a project described as: Karpathy's LLM Wiki implementation plugin for Obsidian - turns notes and PDFs into a linked, LLM-powered knowledge base with entity pages, concept pages, graph-powered Q&A, and local-first privacy.
Dragon-Brain CLAUDE.md
Instructions for iikarus/Dragon-Brain, covering dragon brain — claude.md, the harness, audit remediation (april–may 2026, complete 2026-05-09), the lie this audit closed and the contract that matters now.
zettelkasten-second-memory AGENTS.md
Instructions for cx2002302-lang/zettelkasten-second-memory, covering zettelkasten project guidelines, 系统版本信息（每次思考必须考虑）, 兼容性保障索引, openclaw 2026.6.x 兼容性变更 and openclaw 2026.4.24 配置字段备忘.