skills/nodnarbnitram__claude-code-extensions
1
0
Fork 0
mirror of https://github.com/nodnarbnitram/claude-code-extensions.git synced 2026-09-14 18:36:21 +08:00
Code Issues Packages Projects Releases Wiki Activity
Files
main
nodnarbnitram__claude-code-…/plugins/cce-research
T
History
Brandon Martin 2232aae10c fix(plugins): remove unsupported manifest path fields (#7) 
## Summary
- remove explicit `agents`, `skills`, and `commands` fields from
generated plugin manifests
- rely on Claude Code's standard auto-discovery for plugin-root
`agents/`, `skills/`, and `commands/` directories
- flatten packaged agents to `agents/*.md` so discovery does not depend
on nested-path recursion
- keep the fix minimal by only retaining the explicit `hooks` entry for
`cce-core`

## Why
A local plugin install failed with:

```text
Plugin has an invalid manifest file ... Validation errors: agents: Invalid input
```

Our packaged plugins already follow the standard directory structure, so
the extra manifest path fields were unnecessary and were the most likely
validator mismatch. Greptile also flagged that many generated plugin
agents were nested under paths like `agents/specialized/...`, which
could silently fail if discovery is non-recursive. This change aligns
the packages with the default plugin structure instead of relying on
special manifest fields or recursive discovery.

## Changes
- update `scripts/sync_plugin_packages.py` to stop emitting manifest
path overrides
- flatten generated packaged agents to plugin-root `agents/*.md`
- regenerate all packaged plugin manifests with minimal metadata-only
manifests
- regenerate all packaged plugin agent files into the flat standard
layout

## Verification
- `python3 scripts/sync_plugin_packages.py`
- `python3 -m py_compile scripts/sync_plugin_packages.py
install_extensions.py`
- validated all 19 generated plugin manifests as JSON
- confirmed no generated manifest still contains `agents`, `skills`, or
`commands`
- confirmed packaged agents are flat: `flat_agents=78 nested_agents=0`
2026-04-01 22:25:42 -05:00
..
.claude-plugin
fix(plugins): remove unsupported manifest path fields (#7)
2026-04-01 22:25:42 -05:00
agents
fix(plugins): remove unsupported manifest path fields (#7)
2026-04-01 22:25:42 -05:00
README.md
refactor(plugins): package self-contained plugin installs (#5)
2026-04-01 21:51:29 -05:00

README.md

CCE Research Plugin

Deep research coordination: academic papers, technical analysis, data insights, and web intelligence

Overview

The CCE Research plugin provides a comprehensive suite of specialized research agents that work in coordinated teams to conduct thorough, multi-faceted investigations. Whether you need academic literature reviews, current web intelligence, technical code analysis, or quantitative data insights, this plugin orchestrates expert researchers to deliver comprehensive findings.

Key Features

  • Coordinated Research Team: Strategic orchestration across multiple specialist researchers
  • Multi-Source Intelligence: Academic papers, web content, code repositories, and statistical data
  • Temporal Awareness: All agents track dates and prioritize recent, relevant information
  • Structured Output: JSON-formatted findings with citations, confidence levels, and quality assessments
  • Private Repository Access: Technical researcher can analyze both public and private GitHub repositories

Agents Included

1. Research Coordinator (research-coordinator)

The strategic orchestrator for complex research tasks

  • Analyzes research requirements and allocates tasks to specialists
  • Defines iteration strategies (1-3 rounds) based on complexity
  • Sets quality thresholds and success criteria
  • Plans integration of diverse findings
  • Outputs comprehensive JSON execution plans

Use when: You have a complex research question requiring multiple perspectives or deep investigation.

Capabilities:

  • Strategic task allocation based on specialist strengths
  • Iteration planning for comprehensive coverage
  • Quality assurance and success criteria definition
  • Integration planning for multi-researcher findings

2. Academic Researcher (academic-researcher)

Specialist for scholarly sources and peer-reviewed literature

  • Searches academic databases (ArXiv, PubMed, Google Scholar)
  • Evaluates peer-reviewed papers and research quality
  • Analyzes citations and bibliometric data
  • Identifies research gaps and future directions
  • Extracts research methodologies and limitations

Use proactively for:

  • Research paper analysis
  • Literature reviews
  • Citation tracking
  • Academic methodology evaluation

Output includes:

  • Key findings with confidence levels
  • Research methodology analysis
  • Citation networks and seminal works
  • Quality indicators (journal impact, peer review status)
  • Properly formatted academic citations

3. Web Researcher (web-researcher)

Expert in current news, industry reports, and real-time web intelligence

  • Gathers breaking news and current events
  • Analyzes industry reports and market trends
  • Tracks company announcements and press releases
  • Monitors public sentiment and social discussions
  • Evaluates regulatory updates and policy changes

Use proactively for:

  • Current news and industry reports
  • Real-time information gathering
  • Market trends and competitive intelligence
  • Breaking developments and announcements

Output includes:

  • Executive summary with temporal context
  • Key findings with source citations and dates
  • Source quality analysis (credibility, publication date)
  • Trend identification (current, emerging, long-term)
  • Temporal analysis of recent developments
  • Information gaps and conflicting sources

4. Technical Researcher (technical-researcher)

Specialist in code repositories, technical documentation, and implementation analysis

  • Analyzes GitHub repositories (public and private)
  • Reviews API documentation and technical specs
  • Evaluates code quality and architecture
  • Tracks version histories and breaking changes
  • Assesses community adoption and maintenance status

Use for:

  • Code repository analysis
  • Technical documentation review
  • Implementation examples and best practices
  • Private repository access (using gh CLI and git)

Capabilities:

  • Private repository access: Uses gh CLI for quick file reads, git clone for comprehensive analysis
  • Architecture and design pattern evaluation
  • Security considerations and testing coverage
  • Community activity assessment
  • Version currency and update frequency tracking

Output (JSON format):

  • Repository statistics and metrics
  • Code quality assessment
  • Technical insights and patterns
  • Implementation recommendations
  • Community insights and adoption trends
  • Temporal analysis (maintenance health, ecosystem maturity)

5. Data Analyst (data-analyst)

Quantitative analysis and statistical insights specialist

  • Identifies numerical data from statistical databases
  • Performs comprehensive statistical analysis
  • Creates meaningful comparisons and benchmarks
  • Identifies trends, correlations, and outliers
  • Suggests appropriate data visualizations

Use for:

  • Quantitative analysis
  • Statistical insights
  • Data-driven research
  • Trend analysis and forecasting

Capabilities:

  • Descriptive statistics and trend analysis
  • Comparative benchmarking
  • Correlation analysis and outlier detection
  • Data quality assessment
  • Visualization recommendations

Output (JSON format):

  • Key metrics with confidence levels
  • Trend analysis with rate of change
  • Comparisons with statistical significance
  • Data-driven insights and implications
  • Visualization suggestions
  • Data quality assessment

Installation

Via Plugin Manager

# Add the CCE marketplace (if not already added)
/plugin marketplace add /path/to/claude-code-extensions

# Install the research plugin
/plugin install cce-research@cce-marketplace

Manual Installation

Clone the repository and the agents will be available:

git clone https://github.com/nodnarbnitram/claude-code-extensions.git
cd claude-code-extensions

The agents are in ${CLAUDE_PLUGIN_ROOT}/agents/deep-research/.

Usage

Command Namespacing

When installed as a plugin, agents appear with the prefix cce-research: in the agents list. When used standalone (cloned repository), agents have no prefix.

Installation Mode Agent Names
Plugin cce-research:research-coordinator, cce-research:academic-researcher, etc.
Standalone research-coordinator, academic-researcher, etc.

Example Workflows

1. Complex Research Investigation

> I need comprehensive research on quantum computing's impact on cryptography

Claude will invoke the research-coordinator to:

  1. Analyze the complexity and identify knowledge domains
  2. Allocate tasks to specialists:
    • academic-researcher: Peer-reviewed papers on quantum algorithms
    • web-researcher: Current industry developments and announcements
    • technical-researcher: Quantum computing frameworks and implementations
    • data-analyst: Statistical analysis of algorithm performance
  3. Define integration plan for synthesizing findings

2. Academic Literature Review

> Review recent papers on transformer architecture optimizations

Claude will invoke academic-researcher to:

  • Search academic databases for recent publications
  • Evaluate paper quality and citations
  • Extract key methodologies and findings
  • Identify research gaps

3. Technology Assessment

> Analyze the current state of Rust web frameworks

Claude will invoke technical-researcher to:

  • Search GitHub for popular Rust web frameworks
  • Analyze repository statistics and maintenance status
  • Review documentation and code quality
  • Assess community adoption and trends

4. Market Data Analysis

> What are the growth trends in cloud computing spending?

Claude will invoke data-analyst to:

  • Identify authoritative data sources
  • Extract statistical data
  • Calculate growth rates and trends
  • Provide visualization recommendations

5. Current Events Research

> What are the latest developments in AI regulation?

Claude will invoke web-researcher to:

  • Search for recent news and announcements
  • Track regulatory changes and policy updates
  • Analyze source credibility and recency
  • Identify emerging trends

Key Principles

All research agents follow these core principles:

  1. Temporal Awareness: Always establish current date/time context at the start
  2. Source Quality: Evaluate credibility, recency, and reliability
  3. Structured Output: Consistent JSON formatting with citations
  4. Confidence Levels: Explicit uncertainty acknowledgment
  5. Comprehensive Coverage: Multi-source validation and cross-referencing
  6. Actionable Insights: Focus on decision-relevant findings

Advanced Features

Private Repository Access

The technical-researcher can access private GitHub repositories using:

GitHub CLI (gh):

# View repository details
gh repo view owner/repo

# Read file contents
gh api repos/owner/repo/contents/path/to/file --jq .content | base64 -d

# Search code in private repo
gh search code --repo owner/repo "search term"

Git CLI:

# Clone for comprehensive analysis
git clone https://github.com/owner/repo.git

Multi-Iteration Research

The research-coordinator can plan 1-3 iteration rounds:

  • Single pass: Well-defined, focused topics
  • 2 iterations: Topics requiring initial exploration then deep dive
  • 3 iterations: Complex topics needing discovery, analysis, and synthesis phases

Integration Strategies

The coordinator plans how findings will be synthesized:

  • Complementary: Different aspects of the same topic
  • Comparative: Multiple perspectives on contentious issues
  • Sequential: Building upon each other's findings
  • Validating: Cross-checking facts across sources

Configuration

No additional configuration required. All agents use inherited tools and models from your Claude Code settings.

Tool Access

Research agents use these tools:

  • Read, Write, Edit: File operations for storing findings
  • WebSearch, WebFetch: Web and academic search capabilities
  • Bash: Running date for temporal context, gh/git for repository access
  • Task: For complex multi-step research workflows
  • mcp__context7, mcp__Ref: MCP integrations (if available)

Troubleshooting

Agent Not Triggering

Ensure your prompt includes clear research intent:

  • Use keywords like "research", "analyze", "investigate", "review"
  • Specify the domain (academic, technical, market, current events)

Private Repository Access Fails

For technical-researcher:

  1. Ensure gh CLI is authenticated: gh auth status
  2. Verify repository access permissions
  3. Check network connectivity

Output Format Issues

All agents output structured JSON. If you need different formats:

  • Ask agents to format findings as markdown reports
  • Request specific sections or summaries

Contributing

This plugin is part of the claude-code-extensions project.

To contribute:

  1. Fork the repository
  2. Add/modify agents in ${CLAUDE_PLUGIN_ROOT}/agents/deep-research/
  3. Update plugin.json and README.md
  4. Submit a pull request

License

MIT License - See repository LICENSE file

Version History

  • 1.0.0 (2025-12-23): Initial release
    • 5 specialized research agents
    • Coordinated multi-researcher workflows
    • Temporal awareness and recency tracking
    • Private repository access
    • Structured JSON output formats

Support

Part of the Claude Code Extensions ecosystem - Modular plugins for every tech stack

Reference in New Issue View Git Blame Copy Permalink