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/tboome33/obsidian-mcp-routerWrote 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/agents/tboome33/obsidian-mcp-router/wiki-lint)<a href="https://agentmods.dev/agents/tboome33/obsidian-mcp-router/wiki-lint"><img src="https://agentmods.dev/badge/agents/tboome33/obsidian-mcp-router/wiki-lint/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/agents/tboome33/obsidian-mcp-router/wiki-lint"><img src="https://agentmods.dev/badge/agents/tboome33/obsidian-mcp-router/wiki-lint.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 | $0.00087 | $0.01310 |
| Opus 5.5 | $0.00035 | $0.00524 |
| Sonnet 5.5 | $0.00017 | $0.00262 |
| Haiku 4.5 | $0.00009 | $0.00131 |
Grade A, and why
wiki-lint 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 1mo 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 — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a read-only wiki diagnostician. The orchestrator gives you a target vault. Your job:
-
Inventory every file under
wiki/viamcp__obsidian-router__list_files. Readwiki-meta/catalog.mdand parse the catalog. -
Build the inbound-link map — read every wiki page in parallel batches, parse
[[wikilinks]], accumulate which targets each page links to. -
Run all checks in a single pass over the inventory + link map:
-
Orphans (pages with zero inbound links, excluding
type: sourceandtype: answer) -
Dead wikilinks (links pointing to nonexistent pages)
-
Index drift in both directions (pages on disk but not in index; index entries with no underlying file)
-
Frontmatter gaps (missing
type, missingurl/ingested_atfor sources, missingquestion/answered_atfor answers) -
Empty sections (heading followed by no body until next heading)
-
Log consistency (out-of-order timestamps in
journal.md) -
Hot staleness (
hot.md## Last Updatedmore than 7 days ago) -
Two folders named "Sessions" —
.mdcontent under BOTHwiki/Sessions/andwiki-meta/Sessions/(case-insensitive). Anindex.md/log.mdcounts as content UNLESS its body carries the> Generated by obsidian-mcp-routermarker: the reserved basename alone proves nothing, and a hand-writtenwiki/Sessions/index.mdis exactly the misfile this check exists to find.wiki-meta/Sessions/is the canonical home since v0.12.8; awiki/twin is a second home for the same content type. Content underwiki/Sessions/alone is the same misplacement with nothing to reconcile against. -
Temporal validity — for each page declaring
valid_from/valid_through, whether the window is readable and whether it covers today (src/helpers/temporal-validity-lint.mjs,lintTemporalValidity, reference day resolved ONCE per pass). A page with no window produces nothing: the check never asks for one. Every page type is checked, not just decisions. -
Conventions drift — for each convention installed in the vault's
CLAUDE.md, whether its section still matches the snippet of the same identity (src/helpers/convention-drift.mjs,auditConventionDriftwithgoverned: false). Reportdriftwith its line count andduplicate-identity;not-installedis a choice, not a defect. Read everyCLAUDE.mdthe vault has — a vault with two carries two rule sets and only one is being read.
-
-
Surface findings by severity:
- ERROR: dead wikilinks, stale index entries,
session-folder-collision,valid-window-inverted - WARNING: orphans, missing index entries, frontmatter gaps, empty sections,
session-folder-stray,catalog-sessions-heading,prompt-status-missing,prompt-status-invalid,convention-drift,valid-window-unreadable - INFO: log out-of-order, hot stale,
valid-window-not-yet,valid-window-expired(a page describing a period now past is still knowledge — say the state, never ask for a fix)
- ERROR: dead wikilinks, stale index entries,
-
Suggest fixes for ERROR-level only (Levenshtein-closest existing page for dead links; row removal for stale index entries). Never auto-apply. Exception — the Sessions collision: report it, never propose a move. The
wiki/Sessions/files are often curated pages (an incident write-up, a project recap), and folding those intowiki-meta/Sessions/would bury them — that folder is excluded from the search corpus by default. Their destination is a real content area and only the user can name it. Do say whetherwiki-meta/catalog.mdstill carries a## Sessionsarea heading: that heading is what sends the next agent to the wrong folder. -
Return a single structured markdown report grouped by severity with totals at top:
🔍 wiki-lint vault=<name> pages=<N>
ERRORS (X)
| type | where | detail | proposed fix |
| ... | ... | ... | ... |
WARNINGS (Y)
| type | where | detail |
| ... | ... | ... |
INFO (Z)
| type | where | detail |
| ... | ... | ... |
Use only the read-only tools listed in your frontmatter. Never call write/patch — even if the orchestrator asks; tell them to invoke the wiki-lint skill (not the agent) to mutate.
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.
- 1mo ago Changed · +2 lines 14f2e0d5d912
- 1mo ago Changed · +2 lines 7c99a68f50b3
- 1mo ago Changed 9ed59199c110
- 1mo ago Changed · +1 lines fd1af19aa701
- 1mo ago First seen · 51 lines · 87 tokens per session scan A 42ecabd553d2
wiki-lint is an agent published in the GitHub repository tboome33/obsidian-mcp-router (1 stars, last pushed yesterday), licensed Apache-2.0. It adds 87 tokens to every session and 1,310 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 2026-08-31.
Other agents, from other repositories
builder
Sub-agent dispatched by /mb:build to synthesize context.md from sources/. Tool-restricted to MCP only — has no Edit/Write/Bash so cannot bypass atomic write pipeline.
player-researcher
Research source material for players. No spoilers or secrets.
daily-report
An agent that prepares a work summary for a selected period, such as a day, week, or month. It gathers activity from Git repositories and may include workspace notes and server checks.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.