wiki-research
Perform multi-round web research on a topic, synthesize the findings, and file structured results into the Obsidian wiki. Use for comprehensive web-sourced research intended to become wiki knowledge.
Install / Use
npx skills add Ar9av/obsidian-wiki --skill wiki-researchInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
Content & MediaSupported Platforms
Tags
Our assessment of wiki-research
wiki-research scores 95/100 on our quality scale, 130th of 1,155 Content & Media skills we index (top 12%).
Its SKILL.md is 13 KB long, well organised into 26 sections with 10 code examples: a thorough specification that gives an agent plenty to work with.
With 3,485 GitHub stars, it is one of the more widely adopted skills in the catalogue.
Maintenance, license and trust
- The repository was last updated about 2 months ago, so wiki-research is actively maintained.
- Our last check on 2026-11-16 could not reach the source (HTTP 404); it may have been moved or deleted.
- It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
- Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit โ read the skill file before letting an agent act on it.
wiki-research compared with similar skills
All 4 of these similar skills score higher than wiki-research; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| wiki-research (this skill)by Ar9av | 95 | 3.5k | 54d ago | SKILL.md |
| siyuanby siyuan-note | 100 | 47.3k | today | MCP Server |
| x402by browser-use | 100 | 121.0k | 3d ago | SKILL.md |
| algorithmic-artby anthropics | 100 | 177.9k | 55d ago | SKILL.md |
| pptxby anthropics | 100 | 177.9k | 55d ago | SKILL.md |
Frequently asked questions
- How do I install wiki-research?
- Run
npx skills add Ar9av/obsidian-wiki --skill wiki-research. The install tabs above show the steps for each supported agent. - Which AI agents does wiki-research work with?
- It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is wiki-research safe to use?
- It is MIT-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is wiki-research still maintained?
- The repository was last updated about 2 months ago, so wiki-research is actively maintained.
Skill content
View source on GitHubname: wiki-research description: > Perform multi-round web research on a topic, synthesize the findings, and file structured results into the Obsidian wiki. Use for comprehensive web-sourced research intended to become wiki knowledge.
Wiki Research โ Autonomous Multi-Round Research
You are running an autonomous research loop on a topic, synthesizing what you find, and filing the results into the Obsidian wiki as permanent knowledge.
Before You Start
Writing profile: Before drafting or rewriting natural-language Markdown, read and apply the Writing Profile Resolution section in llm-wiki/SKILL.md. Framework schema, provenance, safety, and operation-specific requirements take precedence.
WRITING.md preferences apply only to newly drafted or rewritten natural-language Markdown; preserve source content and structured records.
- Resolve config โ follow the Config Resolution Protocol in
llm-wiki/SKILL.md(inline@nameoverride โ walk up CWD for.envโ global config โ prompt setup). This givesOBSIDIAN_VAULT_PATHandOBSIDIAN_LINK_FORMAT(default:wikilink). - Read
$OBSIDIAN_VAULT_PATH/index.mdto understand what's already in the wiki โ don't re-research things the wiki covers well - Read
$OBSIDIAN_VAULT_PATH/hot.mdif it exists โ it surfaces recent context - Check
$OBSIDIAN_VAULT_PATH/references/research-config.mdif it exists โ it may define source preferences, domains to skip, or confidence rules for this vault - Check
$OBSIDIAN_VAULT_PATH/references/research-backends.mdif it exists โ it registers optional CLI retrieval backends (social media, video transcripts, paid APIs, etc.). Load any available backends into your working state for this session.
When writing internal links in generated pages, apply the link format from llm-wiki/SKILL.md (Link Format section) using the OBSIDIAN_LINK_FORMAT value.
Confirm the research topic with the user if it's ambiguous. Then proceed.
Research Configuration (optional)
If references/research-config.md exists in the vault, read it and apply any rules it defines:
- Source preferences (e.g., prefer academic sources, avoid certain domains)
- Domains to skip
- Confidence scoring adjustments
- Topic-specific constraints
If the file doesn't exist, proceed with defaults.
Research Backends (optional)
If references/research-backends.md exists in the vault, load it before starting research. It defines zero or more CLI retrieval backends as a YAML list:
backends:
- name: yt-dlp-transcript # friendly label
binary: yt-dlp # CLI binary (checked with `command -v`)
invoke: "yt-dlp --skip-download --write-auto-sub --sub-lang en --sub-format json3 -o /tmp/ytvid '{url}'"
when_to_use: YouTube video URLs, video transcripts
cost_tier: free # free | paid
env_key: "" # required env var for paid tiers (empty = always enabled)
output: text # json | text | markdown
- name: perplexity-sonar
binary: perplexity
invoke: "perplexity search '{query}'"
when_to_use: deep synthesis queries needing multi-source aggregation
cost_tier: paid
env_key: PERPLEXITY_API_KEY # skipped if unset
output: text
Backend availability check (run once at session start):
- For each backend:
command -v <binary> 2>/dev/nullโ if not found, mark unavailable and note it in the run summary - For
paidbackends: also check that$env_keyis non-empty โ if unset, mark unavailable and note it - Build a list of active backends (available + key-gated checks pass) to use in Rounds 1โ2
Invocation rules (per angle/URL during research):
- Substitute
{url}or{query}in theinvoketemplate with the current URL or search query - Capture stdout; on non-zero exit code โ skip this backend for this angle, note the short error, continue
- Fold backend output into the same claims/concepts/entities/contradictions extraction, citing the source URL the backend returns (or the query string for query-mode backends)
- A backend failure never aborts the research run โ always fall back to
WebSearch/WebFetch
Free-first ordering: Evaluate free backends before paid ones for each angle. If a free backend returns sufficient content, paid backends for the same angle can be skipped.
No research-backends.md โ skip this section entirely; behavior is identical to today.
Starter registry template
If the user asks for an example registry, offer this file at $VAULT/references/research-backends.md:
# Optional CLI backends for wiki-research. Delete rows you don't need.
# Skill docs: skills/wiki-research/SKILL.md โ Research Backends section
backends:
# --- free / local ---
- name: defuddle-fetch
binary: defuddle
invoke: "defuddle parse '{url}' --markdown"
when_to_use: any URL โ cleaner extraction than WebFetch alone
cost_tier: free
env_key: ""
output: markdown
- name: yt-dlp-transcript
binary: yt-dlp
invoke: "yt-dlp --skip-download --write-auto-sub --sub-lang en --sub-format json3 -o /tmp/ytvid '{url}'"
when_to_use: YouTube video URLs for transcript extraction
cost_tier: free
env_key: ""
output: text
# --- paid / gated (skipped when env key is unset) ---
- name: perplexity-sonar
binary: perplexity
invoke: "perplexity search '{query}'"
when_to_use: deep synthesis queries needing multi-source aggregation
cost_tier: paid
env_key: PERPLEXITY_API_KEY
output: text
Round 1 โ Broad Survey
Goal: Get a wide map of the topic.
- Decompose the topic into 3-5 distinct angles (e.g., for "vector databases": what they are, when to use them, leading implementations, trade-offs, production gotchas)
- For each angle, run 2-3
WebSearchqueries using varied phrasing - For the top 2-3 results per angle, use
WebFetch(ordefuddle parse <url> --markdownif available โ cleaner extraction) to get content. For each URL, also invoke any active backends whosewhen_to_usematches (e.g., a YouTube URL triggersyt-dlp-transcript); fold their output into extraction alongsideWebFetchresults, citing the source URL the backend returns. - From each fetched page, extract:
- Key claims โ what the source explicitly states
- Concepts โ ideas, terms, frameworks introduced
- Entities โ tools, people, organizations mentioned
- Contradictions โ places where sources disagree with each other
Track what's covered and what's missing as you go.
Round 2 โ Gap Fill
Goal: Close the holes left by Round 1.
Review what Round 1 produced:
- What questions did sources raise but not answer?
- Where do sources contradict each other?
- Which angles got thin coverage?
Run up to 5 targeted searches specifically addressing these gaps. Prefer primary sources, official documentation, and authoritative analyses over link aggregators. For gap-fill queries, also invoke any active query-mode backends (e.g., perplexity-sonar) by substituting {query} in their invoke template โ fold results into extraction with backend name as citation context.
Add findings to your working set. Update the contradiction list.
Round 3 โ Synthesis Check
Goal: Resolve contradictions; confirm depth is sufficient.
If major contradictions remain unresolved:
- Run one final targeted pass (2-3 searches) to find authoritative resolution
- If resolution is impossible, flag the contradiction explicitly in the synthesis page
If contradictions are minor or the topic feels well-covered after Round 2, skip additional searching and proceed to filing.
Halt condition: Stop when depth is achieved or 3 rounds are complete โ do not loop indefinitely.
Filing โ Write Wiki Pages
Organize all findings into wiki pages across four output areas:
1. sources/ โ One page per major reference
For each significant source (typically 4-8 pages total):
---
title: >-
<Source title>
category: references
tags: [<2-4 domain tags>]
sources:
- "<URL>"
source_url: "<URL>"
created: <ISO-8601 timestamp>
updated: <obsidian-wiki memory now>
summary: >-
<1-2 sentences describing what this source covers, โค200 chars>
provenance:
extracted: 0.X
inferred: 0.X
ambiguous: 0.X
base_confidence: <0.17 + 0.5 ร classify(url) for a single source>
lifecycle: draft
lifecycle_changed: <obsidian-wiki memory now --date>
---
Body: title, URL, what it covers, key claims (with provenance markers), limitations.
2. concepts/ โ One page per substantive concept
For each significant concept surfaced across sources:
Standard concept frontmatter + body. Link concepts to each other and to source pages.
3. entities/ โ Tools, organizations, people
For each significant entity encountered (tools, libraries, companies, key authors):
Standard entity frontmatter. Link back to concepts that use the entity and sources where it appears.
4. synthesis/Research: [Topic].md โ Master synthesis
The primary output: a structured synthesis of everything found.
---
title: >-
Research: <Topic>
category: synthesis
tags: [<3-5 domain tags>, research]
sources: [<list of source URLs or page paths>]
created: <ISO-8601 timestamp>
updated: <obsidian-wiki memory now>
summary: >-
Synthesis of <N>-round research on <topic>. Covers <core findings in โค200 chars>.
provenance:
extracted: 0.X
inferred: 0.X
ambiguous: 0.X
base_confidence: <min(N_unique_sources/3,1.0)ร0.5 + avg_source_qualityร0.5>
lifecycle: draft
lifecycle_changed: <obsidian-wiki memory now --date>
---
# Research: <Topic>
## Overview
<2-4 sentence executive summary of what the research found>
## Key Findings
<Bulleted list of the most important claims, each with a [[source page]] citation>
## Core Concepts
<Links to concept pages created, with one-line descriptions>
## Entities & Tools
<Links to entity pages, with one-line descriptions>
## Contradictions & Open Questions
<Where sources disagree or where the research hit limits>
## Sources Consulted
<Linked list of all source pages>
Cross-linking
After filing all pages:
- Every concept page should link to at least 2 source pages
- Every source page should link to the concept pages it informed
- The synthesis page should link to all concept, entity, and source pages produced
Check index.md for existing pages on the same topics โ merge into existing pages rather than creating duplicates.
Update Tracking Files
.manifest.json โ Add a research entry:
{
"type": "research",
"topic": "<topic>",
"researched_at": "TIMESTAMP",
"rounds_completed": 3,
"sources_fetched": N,
"pages_created": ["..."],
"pages_updated": ["..."]
}
One locked call updates the index, the log, and the hot cache:
obsidian-wiki memory sync WIKI_RESEARCH \
topic="<topic>" rounds=<N> sources_fetched=<N> \
pages_created=<M> backends_used="<name,...|none>" \
--takeaways "<the research topic and its core finding, in one line>" --expect-takeaways-sha256 "$TAKEAWAYS_REVISION"
Do not list every page you created in a log field โ memory sync reconciles the index from disk, and a huge field only crowds the hot cache. If the research is ongoing, record it: obsidian-wiki memory todo add "<open question>" --origin synthesis/<page>.md.
Never hand-edit index.md, log.md, or hot.md โ the command takes the lock that keeps a parallel writer from dropping your update.
See skills/llm-wiki/references/MEMORY.md for the full procedure.
Quality Checklist
- [ ] 3 rounds completed (or halted at sufficient depth)
- [ ] Synthesis page exists at
synthesis/Research: [Topic].md - [ ] Source pages written for major references
- [ ] Concept and entity pages written for significant items
- [ ] Contradictions flagged in synthesis page
- [ ] All pages cross-linked
- [ ]
index.md,log.md,hot.md,.manifest.jsonupdated - [ ] Backend summary reported: which ba
Truncated for display โ read the full file on GitHub.
Related Skills
siyuan
47.3kAn open-source, privacy-first, self-hosted knowledge workspace where humans and AI agents work together ๅผๆบใ้็งไผๅ ใ่ชๆ็ฎก็็ฅ่ฏๅทฅไฝ็ฉบ้ด๏ผ่ฎฉไบบไธๆบ่ฝไฝๅจๆญคๅไฝ
x402
121.0kSet up Browser Use Cloud payments with x402 โ pay per request from a crypto wallet (USDC on Base mainnet), no signup or API key.
algorithmic-art
177.9kCreating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems.
pptx
177.9kUse this skill any time a .pptx or .potx file is involved in any way โ as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx or .potx file (even if the extracted content will be used elsewhere, like in an emโฆ
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit โ see the Safety scan above for what the skill file itself contains.
