Research
Run structured multi-wave agent research before you commit to a stack, vendor, architecture, or competitive positioning.
Overview
Research is a journey-wide agent skill that runs wave-based multi-agent knowledge gathering with deferred synthesis—usable whenever a solo builder needs evidence across sources before committing to a technology, architec
Install
npx skills add https://github.com/hyperb1iss/hyperskills --skill researchWhat is this skill?
- Wave-based breadth-first gathering with deferred synthesis so conclusions are not locked on the first few hits
- Five-step shape: Prime → Wave 1 broad sweep → Gap analysis → targeted follow-up waves → Synthesize
- Calibrated depth: quick mode, standard one-wave-plus-follow-ups, or full deep-dive pattern
- Designed for technology evaluation, codebase archaeology, competitive analysis, and compare-alternatives questions
- Pattern distilled from 300+ real research dispatches focused on actionable intelligence
- Five-step research shape: Prime, Wave 1, Gap analysis, Wave 2+, Synthesize
- Pattern mined from 300+ real research dispatches
Adoption & trust: 590 installs on skills.sh; 13 GitHub stars; 2/3 security scanners passed (skills.sh audits).
What problem does it solve?
You need a recommendation backed by more than the first page of search results, but ad-hoc browsing locks in conclusions before you have seen the full landscape.
Who is it for?
Technology comparisons, SOTA checks, competitive scans, and codebase archaeology when the question clearly benefits from multiple waves and gap-driven follow-ups.
Skip if: A single known doc lookup, a trivial config fact you already have a link for, or cases where you already have an approved spec and only need execution—not more discovery.
When should I use this skill?
Gathering knowledge at scale before decisions—technology evaluation, SOTA analysis, codebase archaeology, competitive analysis—or when the user mentions research, investigate, evaluate options, compare alternatives, stat
What do I get? / Deliverables
You get an accumulated evidence base from staged broad and targeted research waves, then a sharper synthesis and recommendation instead of a premature single-pass answer.
- Accumulated findings across research waves
- Gap analysis notes driving follow-up waves
- Final synthesis with actionable recommendations
Recommended Skills
Journey fit
Useful at every journey phase - explore requirements and options before committing to a direction.
Where it fits
Map competitors and positioning before you commit to a niche or feature set.
Compare prototype approaches and trim scope using a gap-driven second research wave.
Evaluate payment, auth, or hosting alternatives with broad sweep then targeted deep dives.
Research channel options and state of the art for SEO or community-led launch tactics.
Investigate incident patterns or observability tooling before changing production setup.
How it compares
Use instead of one-shot web search or unstructured agent rambling when the decision warrants breadth-first gathering and explicit gap analysis before synthesis.
Common Questions / FAQ
Who is research for?
Solo and indie builders using Claude Code, Cursor, Codex, or similar agents who make stack, vendor, and product decisions without a dedicated research team.
When should I use research?
At idea/research for competitive and landscape work; during validate/scope when narrowing options; in build/integrations when evaluating libraries or patterns; before ship or launch when comparing distribution or infra choices; and anytime you say investigate, compare alternative
Is research safe to install?
Review the Security Audits panel on this Prism page and your agent’s tool permissions before runs that spawn sub-agents or use network access.
SKILL.md
READMESKILL.md - Research
# Multi-Agent Research Wave-based knowledge gathering with deferred synthesis. Mined from 300+ real research dispatches: the pattern that consistently produces actionable intelligence. **Core insight:** Research breadth-first, synthesize after. Conclusions drawn from the first three results miss nuance the fourth wave would have surfaced. Deploying agents in waves and accumulating findings before synthesizing produces sharper recommendations. **How to read this skill:** the wave structure below is a shape, not a procedure. Quick mode skips most of it. Standard research uses one wave plus targeted follow-ups. Deep dives genuinely need the full pattern. Calibrate to the question, not the framework. ## The Shape ```dot digraph research { rankdir=TB; node [shape=box]; "1. PRIME" [style=filled, fillcolor="#e8e8ff"]; "2. WAVE 1: Broad Sweep" [style=filled, fillcolor="#ffe8e8"]; "3. GAP ANALYSIS" [style=filled, fillcolor="#fff8e0"]; "4. WAVE 2+: Targeted" [style=filled, fillcolor="#ffe8e8"]; "5. SYNTHESIZE" [style=filled, fillcolor="#e8ffe8"]; "6. DECIDE & RECORD" [style=filled, fillcolor="#e8e8ff"]; "1. PRIME" -> "2. WAVE 1: Broad Sweep"; "2. WAVE 1: Broad Sweep" -> "3. GAP ANALYSIS"; "3. GAP ANALYSIS" -> "4. WAVE 2+: Targeted"; "4. WAVE 2+: Targeted" -> "3. GAP ANALYSIS" [label="still gaps", style=dashed]; "3. GAP ANALYSIS" -> "5. SYNTHESIZE" [label="coverage sufficient"]; "5. SYNTHESIZE" -> "6. DECIDE & RECORD"; } ``` --- ## Phase 1: PRIME Lean on existing knowledge before spawning agents. Re-running research that already lives in Sibyl burns tokens and produces duplicate entries. ### Common moves - **Search Sibyl first:** `sibyl search "<research topic>"`, `sibyl search "<related technology>"`, `sibyl search "<prior decision in this area>"`. Surface what's already known before generating new findings. - **Check for staleness.** Fast-moving topics (frameworks, models, cloud services) usually warrant re-research even when Sibyl has recent entries; treat the existing knowledge as a baseline. Stable topics with recent entries often don't need a fresh pass at all. - **Sharpen the research question.** "Research databases" is too vague to dispatch on. "Compare PostgreSQL vs CockroachDB for multi-region write-heavy workloads with <10ms p99 latency" gives agents enough scope to do useful work. - **Calibrate the research budget** to the decision the research is feeding: | Depth | Agents | Time | When | | -------------- | ------ | --------- | -------------------------------------------- | | **Quick scan** | 2-3 | 2-5 min | Known domain, just need latest info | | **Standard** | 5-10 | 10-15 min | Technology evaluation, architecture options | | **Deep dive** | 10-30 | 20-40 min | Greenfield decisions, SOTA analysis | | **Exhaustive** | 30-60+ | 40-90 min | New project inception, competitive landscape | ### Source quality contract This bit is non-negotiable: the value of research collapses when claims rest on stale blog posts. Specific claim types deserve specific source standards: | Claim type | Preferred source | | ----------------------- | ------------------------------------------------ | | Current version | Package registry, release page, or official CLI | | CLI flags / config keys | Official docs or local `--help` output | | Security frameworks | OWASP, NIST, SLSA/OpenSSF, CIS, ISO, PCI sour