Skip to content
SkillModel-led1,200 tokens in use

Wiki Query

Answers a question from your Obsidian wiki with citations and evidence labels, in quick, standard or deep mode, and never changes the vault.

16,653 stars on the repository MIT LicenseAgriciDaniel/claude-obsidian · skills/wiki-query
claude plugin marketplace add AgriciDaniel/claude-obsidian claude plugin install claude-obsidian@agricidaniel-claude-obsidian
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing Wiki Query costs

74tokens idle
1,200tokens in use

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

shelldetected

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Good
  • Frontmatter12/15
  • Structure19/25
  • Examples9/15
  • Output10/15
  • Size10/15
  • Resources10/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent
  • Runs shell commands−5

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 21

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity84
  • Maintenance100
  • Community56
  • Maturity28
  • Momentum40

Wiki Query is the read-only question skill of the claude-obsidian plugin. When you name the vault as the source, the agent reads the hot page and index, retrieves relevant pages and answers with links to the notes it used.

It labels claims by their evidence status, names the gaps it finds without filling them from model memory, and treats everything it reads in the vault as data to cite, never as instructions.

What You Get

  • A direct answer followed by the evidence and caveats needed to use it
  • Citations as wikilinks to the specific page or heading, with the underlying source where one exists
  • Claims labeled accepted, provisional, contested or unsupported, based on the vault’s claim and source ledgers
  • A note of which fallback was used when retrieval was unavailable or stale
  • A named list of missing evidence when the vault cannot answer

Common Use Cases

  • Asking what your notes say about a topic with sources attached
  • Summarizing a corner of a research vault before writing
  • Checking whether a claim in your notes is supported or contested
  • Running a deep review of competing pages on one question

What Makes It Unique

The skill separates what the vault shows from what the agent infers, and it reports weak or conflicting evidence without picking a winner. Read budgets keep a quick lookup small and a deep one bounded. It tells the agent to treat embedded commands in vault pages as untrusted, which matters when notes include pasted web content.

Under The Hood

The skill loads its SKILL.md and reads vault files: the hot page, the index, claim and source ledgers and the pages it cites. For Standard and Deep it runs shell commands that call Python scripts from the installed plugin, one to check that retrieval is verified and one to query the local search index. These commands only read. It does not fetch web pages or call APIs, and it writes no files. If the user wants to keep an answer, it hands the answer to the separate save skill.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers