Research and vote
Research a topic using multiple sources and conduct multi-agent voting. Use when making architectural decisions, choosing dependencies, or designing APIs. Triggers on "research", "decide", "vote on", "consensus".From its SKILL.md
npx -y skills add nexus-substrate/nexus-agents --skill research-and-voteAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 23 stars23 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
3.6 KB, as published. Nobody here has run it
Research and Vote Skill
<!-- CANONICAL SOURCES: - docs/architecture/CONSENSUS_PROTOCOLS.md - docs/research/CONTRIBUTING.md - CLAUDE.md Consensus Voting Protocol - skills/references/orchestration-patterns.md (consensus, fan-out, retry policies) -->Full documentation:
Process
Phase 1: Research
- Check research registry first:
grep -ri "keyword" docs/research/ - Gather primary sources (official docs, specs, RFCs)
- Document findings with source links
Phase 2: Proposal
Create proposal with:
- Problem statement
- Proposed solution
- Alternatives considered
- Trade-offs
Phase 3: Voting
Voting Agents: Architect, Security, DevEx, AI/ML, PM
Thresholds:
| Decision Type | Threshold |
|---|---|
| Reversible changes | Majority |
| Architecture | Supermajority |
| Security-critical | Unanimous |
See CONSENSUS_PROTOCOLS.md for protocol selection matrix.
Phase 4: Documentation
Record decision in GitHub issue with voting record.
Output Format
# Decision Record: [Topic]
## Status: [Approved/Rejected]
## Voting Record
| Agent | Vote | Reasoning |
| ----- | ---- | --------- |
Anti-rationalization — Research and vote
| Excuse | Counter |
|---|---|
| "I already know the answer" | Then write up the alternatives anyway. The vote isn't to discover the answer; it's to surface what you missed. |
| "Skip the research, just vote" | A vote without research surfaces opinion, not informed judgment. Cite primary sources. |
| "Simulated votes are fine for this" | Per CLAUDE.md and memory: simulated votes are random. Never use for real decisions. |
| "Unanimous would be too slow" | Unanimous applies to security-critical and breaking-API. Match the threshold to the reversibility. |
| "We can revisit if it's wrong" | Some decisions are expensive to reverse (data shape, public API, dep choice). Apply higher threshold accordingly. |
Red flags
- Proposal lists no alternatives or only "do X / do nothing"
- Voter reasoning is one sentence ("approve") with no specific cite
simulateVotes: truefor anything other than unit tests- Vote happens before research is done
- Decision recorded but no GitHub issue or commit links the vote
Verification checklist
- Research cites primary sources (official docs, RFCs, papers — not blog posts only)
- At least 2 alternatives considered with why-rejected reasoning
-
consensus_voteran with real CLIs (simulateVotes: false) - Threshold matches decision type (majority / supermajority / unanimous)
- Decision Record posted as a GitHub issue or comment with voter cites
- Voter reasoning includes confidence and specific cites — no rubber-stamps