Comprehensive research and content extraction with 4 depth modes: Quick (1 Perplexity agent, ~10-15s), Standard (4 agents — Claude + Gemini + Grok + Perplexity, cross-checked, ~30-60s), Extensive (7 explorers + 2 independent verifiers, ~60-90s), Deep Investigation (progressive iteration with persistent MEMORY/RESEARCH/ vault, ~3-60min).
A generic AI doing research is one model querying its own training data, hallucinating URLs it can't verify, and returning a single confident perspective on a topic that has real disagreement. There's no cross-checking, no adversarial verification, no distinction between 'this is well-established' and 'this is one source's claim.' When you ask it to find what fans thought of something, you get what press releases said — because it can't tell the difference. Broken links get served as citations.
Four depth modes scale from a single Perplexity agent (Quick, ~10s) up to seven parallel explorers plus two independent verifiers who never see the explorers' reasoning (Extensive, ~60-90s) and a loop-compatible Deep Investigation that builds a persistent vault in MEMORY/RESEARCH/ across sessions. Every URL is verified before delivery — a hallucinated link is a catastrophic failure, not an acceptable approximation. Confidence tagging ([HIGH]/[MED]/[LOW]/[CONFLICT]) appears inline so you know exactly what to trust. When the question is about community sentiment, the skill detects that signal at Step 0 via SourceRoutingProtocol and routes to Reddit JSON API and X before touching recap journalism. Fabric's 242+ patterns slot in for structured extraction. The whole stack feeds into blogging, newsletter, and social post workflows, so research findings move directly into published content without a copy-paste step.
What you say to your DA, and what the Research skill actually does.
The thinking, frameworks, and architecture that distinguish this skill from a generic version of the same task.
Researches a topic across multiple sources and verifies every claim before delivery. Four depth modes scale from a single fast lookup to a multi-session investigation: Quick (1 agent, ~10-15s), Standard (4 agents cross-checked, ~30-60s), Extensive (7 explorers + 2 independent verifiers, ~60-90s), and Deep Investigation (progressive iteration with a persistent vault, ~3-60min). Output is confidence-tagged: [HIGH] [MED] [LOW] [CONFLICT].
A single AI agent doing research has two failure modes that quietly wreck the result. It hallucinates URLs — confident links that go nowhere, which destroys trust in the whole report. And it answers from one angle, so it parrots whatever the first few search results said and misses conflicts, gaps, and what real people actually thought. Recap journalism is the worst offender: ask "what did fans think of X" and a lone agent hands back promoter copy dressed as consensus. This skill runs several agents in parallel, cross-checks and independently verifies their findings, checks every URL before it ships, and routes sentiment questions to community sources first.
Multiple agents work in parallel and their findings get reconciled. Verification runs in three layers at zero added latency: each agent self-verifies its own URLs, a synthesis step cross-checks for conflicts, and dedicated verifier agents (Extensive/Deep) check findings with no access to the explorers' reasoning. Step 0 of every workflow routes sentiment questions to community scrapers before web search, and every URL is verified before delivery — a hallucinated link is a catastrophic failure.
READ: QuickReference.md for detailed examples and mode comparison.
| Trigger | Mode | Speed |
|---|---|---|
| "quick research" | 1 Claude agent | ~10-15s |
| "do research" | 2 agents + cross-check | ~15-30s |
| "extensive research" | 7 explorers + 2 verifiers | ~60-90s |
| "deep investigation" | Progressive iteration + verification | ~3-60min |
Inspired by Nomad (arXiv:2603.29353). Three layers of verification, zero added latency:
| Layer | What | Where | Cost |
|---|---|---|---|
| Self-Verification | Each agent verifies own URLs and tags confidence before returning | All agents | 0s (inside parallel window) |
| Cross-Check | Synthesis step detects conflicts and cross-references findings | Standard, Extensive, Deep | 2-3s (within synthesis) |
| Independent Verification | Dedicated verifier agents with no access to explorer reasoning | Extensive, Deep only | 0s (parallel with explorers) |
Confidence tags in output: [HIGH] [MED] [LOW] [CONFLICT]
See Workflows/Verify.md for full verification protocol.
Progressive iterative research that builds a persistent knowledge vault. Works in both single-run (one cycle) and loop mode (Algorithm-driven iterations).
Concept: Broad landscape → discover entities → score importance/effort → deep-dive one at a time → loop until coverage complete.
Domain template packs customize the investigation for specific domains:
Templates/MarketResearch.md — Companies, Products, People, Technologies, Trends, InvestorsTemplates/ThreatLandscape.md — Threat Actors, Campaigns, TTPs, Vulnerabilities, Tools, DefendersExample invocation:
"Do a deep investigation of the AI agent market"
→ Loads MarketResearch.md template
→ Iteration 1: Broad landscape + first entity deep-dive
→ Loop mode: Each iteration deep-dives the next highest-priority entity
→ Exit: When all CRITICAL/HIGH entities researched + all categories covered
Artifacts persist at ~/.claude/PAI/MEMORY/RESEARCH/{date}_{topic}/ — the vault survives across sessions.
See Workflows/DeepInvestigation.md for full workflow details.
Working files (temporary work artifacts): ~/.claude/PAI/MEMORY/WORK/{current_work}/
~/.claude/ to get the work_dir valueHistory (permanent): ~/.claude/History/research/YYYY-MM/YYYY-MM-DD_[topic]/
_X skill) + YouTube first per SourceRoutingProtocol.md. Recap web search is the secondary source for community-sentiment questions, not the primary one. Quick mode can return recap-only and miss the actual fan data — pull Reddit directly rather than waiting to be asked again. Do not repeat._X, YouTube Data API v3 if YOUTUBE_API_KEY is set) before reaching for Apify or BrightData. Scrapers are fallback for when the API path is unavailable, rate-limited, or doesn't expose the data shape needed (e.g., YouTube transcripts — use fabric -y even when the Data API key is set). The cascade inversion is the recurring failure mode. See SourceRoutingProtocol.md Cascade Priority section for the per-platform table..json to any thread or listing URL. Set User-Agent: PAI-Research/1.0 or Reddit rate-limits the default UA. Apify Reddit scraper is Tier 2 (fallback), not Tier 1.fabric -y URL directly — don't try to scrape YouTube pages with WebFetch.Example 1: Quick lookup
User: "quick research on Hono SSR middleware patterns"
→ Invokes QuickResearch workflow (1 Claude agent)
→ Returns summary with key patterns and links
→ ~10-15 seconds
Example 2: Standard multi-source research
User: "research the current state of AI agent frameworks"
→ Invokes StandardResearch workflow (2 agents: Claude + Gemini, cross-checked)
→ Cross-references findings, confidence-tags, verifies URLs
→ Returns synthesized report with citations
→ ~15-30 seconds
Example 3: Deep investigation
User: "do a deep investigation of the AI agent market"
→ Invokes DeepInvestigation workflow
→ Broad landscape scan → entity discovery → priority scoring → deep-dives
→ Builds persistent knowledge vault in MEMORY/RESEARCH/
→ Loop-compatible for multi-session investigation
Workflows/"research" / "do research" / "research this".md research / do research / research this
Workflows/"quick research" / "minor research".md quick research / minor research
Workflows/"extensive research" / "deep research".md extensive research / deep research
Say any of these to your DA and PAI activates the Research skill automatically:
Or invoke explicitly:
Skill("Research")Auxiliary files the skill loads at runtime — frameworks, guides, configs.
The thinkers, books, frameworks, and research this skill is built on. The ideas belong to them — the integration belongs to PAI.
Install PAI on your machine — your DA gets the Research skill plus 44 others, all hooked into one Life OS.