OXI-717/ai-native-toolkit / plugins/research-ensemble/skills/research-ensemble
research-ensemble
VerifiedMulti-agent adversarial research with Exa AI. Two levels: standard (3-5 agents, 30-60 min) and deep (10-15 agents, full ensemble with 3 cycles). Triggers: "research X", "исследуй", "deep research", "find out about", "compare options for", "deep dive into", "gather information on", "write a report about", "проанализируй", "разберись в", "investigate", "analyze topic", "сравни варианты", "изучи тему"
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 6.4 KB |
| references/agent-prompts/critic.md | markdown | 2.8 KB |
| references/agent-prompts/deep-diver.md | markdown | 2.5 KB |
| references/agent-prompts/domain-reviewer.md | markdown | 2.8 KB |
| references/agent-prompts/fact-checker.md | markdown | 2.7 KB |
| references/agent-prompts/interaction-mapper.md | markdown | 3.4 KB |
| references/agent-prompts/scout.md | markdown | 4.4 KB |
| references/agent-prompts/statistician.md | markdown | 2.5 KB |
| references/agent-prompts/synthesizer.md | markdown | 3.0 KB |
| references/deep-workflow.md | markdown | 9.9 KB |
| references/domains.md | markdown | 6.2 KB |
| references/exa-search-guide.md | markdown | 4.3 KB |
| references/generate-pdf-report.pyexecutable | script | 19.0 KB |
| references/profile-template.md | markdown | 1.6 KB |
| references/standard-workflow.md | markdown | 5.9 KB |
references/exa-search-guide.md
View on GitHub# Exa AI Search Guide
## Overview
Exa is the PRIMARY search tool for all research agents. WebSearch is a fallback only when Exa doesn't return sufficient results.
Exa returns clean, LLM-ready content — no ads, navigation, or boilerplate. This makes it ideal for autonomous subagents that need to process search results without human curation.
## Three Exa Tools
### 1. `mcp__exa__web_search_exa` — Primary Discovery
**When:** Default search for articles, studies, documentation, blog posts, news.
**Key Parameters:**
- `query`: Search query (natural language works best)
- `numResults`: Number of results (default 8, use 10-12 for broad discovery)
- `livecrawl`: Set to `"preferred"` for topics needing fresh 2025-2026 data
**Examples:**
```
# Broad discovery
query: "kubernetes vs nomad orchestration comparison 2026"
numResults: 10
# Specific data
query: "creatine monohydrate dosage meta-analysis RCT"
numResults: 12
# Fresh data
query: "LLM inference cost optimization 2026"
numResults: 8
livecrawl: "preferred"
```
### 2. `mcp__exa__company_research_exa` — Company Intelligence
**When:** Business/product domain — researching companies, competitors, market position.
**Key Parameters:**
- `companyName`: Company name
**Examples:**
```
companyName: "Hetzner"
companyName: "Vercel"
```
### 3. `mcp__exa__get_code_context_exa` — Code & Documentation
**When:** Tech domain — APIs, libraries, implementation patterns, GitHub repos.
**Key Parameters:**
- `query`: What to search for
- `tokensNum`: Amount of context (1000 for quick reference, 5000-10000 for detailed examples, up to 50000 for comprehensive)
**Examples:**
```
query: "fastapi middleware authentication pattern"
tokensNum: 5000
query: "terraform aws ecs module best practices"
tokensNum: 10000
```
## Domain → Tools Mapping
| Domain | Primary Exa Tool | Secondary | WebFetch For |
|--------|-----------------|-----------|--------------|
| **tech** | web_search + code_context | — | Official docs, changelogs |
| **business** | web_search + company_research | — | Annual reports, press releases |
| **health** | web_search | — | PubMed abstracts, clinical trials |
| **security** | web_search + code_context | — | CVE details, security advisories |
| **general** | web_search | — | Specific pages from results |
## Search Strategy for Scouts
Each Scout agent should follow this search pattern:
### Step 1: Broad Discovery (8-12 results)
```
mcp__exa__web_search_exa:
query: "[topic] comprehensive analysis"
numResults: 10
```
### Step 2: Deep Extraction (if needed)
For specific pages that need full content:
```
WebFetch:
url: "[URL from Exa results]"
prompt: "Extract key findings, data points, and conclusions about [topic]"
```
### Step 3: Supplementary Search (if Exa insufficient)
```
WebSearch:
query: "[refined query for gaps]"
```
### Step 4: Fresh Data (if topic is current)
```
mcp__exa__web_search_exa:
query: "[topic] 2025 2026"
livecrawl: "preferred"
```
## Query Formulation Tips
1. **Vary query structure** — don't repeat similar queries:
- Broad: "topic overview analysis"
- Specific: "topic statistics data report 2026"
- Expert: "topic expert opinion research findings"
- Counter: "topic criticism problems limitations"
2. **Use date qualifiers** for freshness:
- "topic 2025 2026" forces recent results
- `livecrawl: "preferred"` gets latest crawled pages
3. **Search for counterarguments explicitly:**
- "topic criticism"
- "topic problems limitations"
- "topic alternative approach"
4. **Track URLs** — don't search the same content twice
## Source Priority
When evaluating sources from Exa results, prioritize:
1. Government sources (.gov) — highest authority
2. Academic institutions (.edu) — research quality
3. Peer-reviewed journals — verified methodology
4. Established news organizations — editorial standards
5. Industry reports from recognized firms — domain expertise
6. Community resources (Wikipedia, StackOverflow) — overview and context only
## Anti-Patterns
- Don't search the same query twice with minor variations
- Don't over-rely on a single source for multiple claims
- Don't ignore sources that contradict your emerging thesis
- Don't skip WebFetch when a source needs deeper extraction
- Don't use more than 15 Exa calls per scout — diminishing returns
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |