Skip to content
SkillModel-led3,231 tokens in use

wiki-research

Runs up to three rounds of web research on a topic and files sources, concepts, entities and a master synthesis into an Obsidian wiki with provenance and open contradictions.

3,792 stars on the repository MIT LicenseAr9av/obsidian-wiki · skills/wiki-research
pip install obsidian-wiki
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing wiki-research costs

41tokens idle
3,231tokens 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
  • Frontmatter8/15
  • Structure17/25
  • Examples15/15
  • Output15/15
  • 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 266

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Excellent
  • Development activity100
  • Maintenance100
  • Community60
  • Maturity28
  • Momentum40

wiki-research runs an autonomous research loop and keeps the result as permanent wiki knowledge. Round one maps the topic from three to five angles, round two fills the gaps, and round three resolves contradictions or flags them, then it stops.

It files source pages, concept pages, entity pages and a synthesis page named Research followed by the topic, links them together and records the run in the manifest, index, log and hot cache.

What You Get

  • Four to eight source pages with key claims and limitations
  • Concept and entity pages linked to the sources that informed them
  • A synthesis page with findings, concepts, entities, contradictions and sources consulted
  • A manifest entry with rounds completed, sources fetched and pages created
  • A backend summary showing which tools were used or skipped, and why

Common Use Cases

  • Researching an unfamiliar technology and keeping a cited record of it
  • Building a topic cluster in a wiki from web sources in one session
  • Adding video transcripts or paid search to research through a backend registry

What Makes It Unique

It has a fixed halt condition of three rounds and a gap-fill round aimed at what round one left open. Optional backends are tried free first and are skipped when their binary or key is missing, and a failure in one never ends the run. Contradictions are listed in the synthesis page and not smoothed over, and new pages merge into existing ones when the index shows the topic is already covered.

Under The Hood

The skill is instructions for the agent. It searches the web and fetches pages with the agent’s web search and fetch tools, or with the defuddle command when installed. Backends listed in the optional registry run as shell commands with the url or query filled in, for example a YouTube transcript tool or a paid search command that needs an API key. It writes markdown pages and the manifest, runs the obsidian-wiki command line tool for timestamps and a locked memory sync, and can run the qmd tool to refresh the search index.

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