# Research Topics Across Web, Docs, and Code

Research questions often require current web evidence, local documentation, and code examples. This skill coordinates those sources and synthesizes findings with clear confidence labels.

## Install

```bash
npx skillstore add consiliency/research
```

## Metadata

- Status: approved
- Slug: consiliency-research
- Skillstore revision: r2
- Version status: missing
- Tree hash: be48c2e15fa28bf1d668853110cd24523c69dc3788f4d0f60d6296f9a972f004
- Author: Consiliency
- GitHub username: Consiliency
- License: MIT
- Repository: https://github.com/Consiliency/treesitter-chunker/tree/main/.ai-dev-kit/skills/research
- Ref: c43861a65bb95efcae259cd161c9d6f4dc7eec6f
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands
- Quality score: 78
- Quality tier: bronze
- Public page: https://skillstore.pages.dev/skills/consiliency-research
- Manifest: https://skillstore.pages.dev/api/skills/consiliency-research/manifest

## Capabilities

- Selects quick, standard, or extensive modes based on speed and depth needs.
- Coordinates web, documentation, and code researchers in parallel.
- Cross-validates findings across source types in standard and extensive modes.
- Labels synthesized findings as high, medium, low, or conflicting confidence.
- Returns source links, local file references, timing, and agent response metadata.

## Use Cases

- Evaluate a Technical Approach: Compare official guidance, local documentation, and implementation examples before selecting a library or architecture.
- Build Evidence-Based Documentation: Gather current sources, code references, and conflicting viewpoints before drafting technical documentation.
- Prepare a Decision Brief: Research several options, assess source agreement, and produce an actionable recommendation with evidence gaps.

## Prompt Templates

### Quick Fact Check

```
Use quick mode to verify [claim]. Cite the strongest web and documentation sources, then state one confidence level.
```

### Standard Technical Research

```
Use standard mode to research [technical topic]. Compare web guidance with local documentation and code examples. Highlight disagreements.
```

### Extensive Decision Comparison

```
Use extensive mode to compare [option A] and [option B] for [context]. Evaluate architecture, implementation, risks, costs, and source conflicts.
```

### Evidence-Led Research Plan

```
Research [complex question] in extensive mode. Assign distinct source angles, cross-validate every major claim, identify evidence gaps, and recommend further research.
```

## Limitations

- Web research requires network connectivity and cannot access paywalled content.
- Quick mode uses limited queries and may miss nuanced information.
- Extensive mode uses more model calls and can take up to ten minutes.
- The referenced synthesis-patterns.md guide is not included in the package.

## Best Practices

- Choose the smallest mode that provides enough source coverage.
- Give the query a clear scope, date range, and decision context.
- Review low-confidence and conflicting findings before using recommendations.

## Anti Patterns

- Do not treat a single source as settled evidence.
- Do not use extensive mode for simple definitions or routine lookups.
- Do not present incomplete results without noting timeouts or missing sources.

## Security Audit

- Audited at: 2026-07-23T12:30:51.927\+00:00
- Summary: All 11 static findings are false positives caused by Markdown code fences, inline paths, and file references. The skill requests web and local searches, but contains no executable commands, credential handling, prompt injection, or exfiltration instructions.

## Stats

- Views: 244
- Downloads: 11
- Favorites: 0
- Popularity score: 0