Help us improve
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
By maxamillion
15-phase OSS framework evaluation methodology with AI bias prevention. Evaluates open source projects for adoption decisions with rigorous safeguards against common AI analysis pitfalls.
npx claudepluginhub joshuarweaver/cascade-code-general-misc-1 --plugin maxamillion-claude-oss-eval-pluginPhase 2: Perform detailed analysis of each candidate identified in Phase 1, focusing on core capabilities, architecture, and initial fit assessment.
Phase 7: Conduct deep technical analysis of candidate architectures including code-level examination and scalability characteristics.
Phase 5: Assess the health, sustainability, and responsiveness of each candidate's open source community for long-term viability.
Phase 13: Introduce specific product context and constraints to ground the evaluation in real business needs.
Phase 11: Quantify total cost of ownership (TCO) including infrastructure, licensing, maintenance, and hidden costs.
An isolated sub-agent that provides unbiased adversarial review of OSS evaluation recommendations. This agent operates WITHOUT access to the full evaluation context to ensure fresh, unbiased perspective.
A specialized sub-agent focused on web search discovery of OSS candidates. This agent handles the research-intensive Phase 1 discovery process with systematic search patterns and source validation.
Core safeguards against common AI analysis pitfalls during OSS framework evaluation. This skill implements the 12 pitfall prevention rules to ensure objective, accurate, and defensible evaluations.
Systematic cross-platform feature verification to prevent false differentiation and ensure accurate feature matrices. This skill provides protocols for verifying that features claimed as "missing" aren't simply named differently.
Ensure accuracy and recency of information sources during OSS evaluation. This skill provides protocols for verifying source credibility, data freshness, and multi-source validation.
Uses power tools
Uses Bash, Write, or Edit tools
Share bugs, ideas, or general feedback.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Use this agent when evaluating new development tools, frameworks, or services for the studio. This agent specializes in rapid tool assessment, comparative analysis, and making recommendations that align with the 6-day development cycle philosophy. Examples:\n\n<example>\nContext: Considering a new framework or library
Developer workflow tools: community scanning, technical decision-making
Architecture analysis, brainstorming, ADR generation, LLM-as-judge comparison, and spec gap resolution.
Verification-first engineering toolkit for Claude Code. 15 skills across a 5-phase spine (Investigate → Design → Implement → Verify → Ship), 8 specialist agents, an interactive setup wizard. Every skill has rationalizations + evidence requirements. Built for senior ICs and tech leads.
Personal Claude Code + Codex dev stack: security hooks, AI-first code conventions, /security-review, /repo-map, /stack-check, portable statusline. Designed to complement other skills-based plugins, not replace them.
Live codebase visualization and structural quality gate — 14 health dimensions graded A-F, dependency analysis, and architecture governance via MCP
Analyzes CVEs in RHOAI releases by querying Pyxis and Red Hat Security APIs
Evaluates open source projects against 30 Key Performance Indicators using multi-agent research with parallel subagents for bias-free assessment
Evaluates contributions to open source projects using LDAP org traversal, GitHub organization membership, or both as roster sources
A comprehensive Claude Code plugin implementing a 15-phase methodology for evaluating open source software frameworks with built-in safeguards against AI analysis pitfalls.
This plugin is a work in progress and is subject to change.
The methodology used here to evaluate OSS frameworks was created by @jwforres and @n1hility, and adapted here as a Claude Code Plugin.
This plugin provides a structured, bias-resistant approach to evaluating OSS frameworks for adoption decisions. It addresses common AI analysis problems including:
[OSS]/[PAID] annotationsInstall directly within Claude Code using the plugin system:
# Add the marketplace (one time)
/plugin marketplace add maxamillion/claude-oss-eval-plugin
# Install the plugin
/plugin install oss-eval
The plugin will be available immediately. Commands are namespaced as /oss-eval:command.
For testing or contributing to the plugin:
# Clone the repository
git clone https://github.com/maxamillion/claude-oss-eval-plugin.git
# Run Claude Code with the plugin directory
claude --plugin-dir /path/to/claude-oss-eval-plugin
Share the plugin with your team via git:
# Add marketplace and install with project scope
/plugin marketplace add maxamillion/claude-oss-eval-plugin
/plugin install oss-eval --scope project
This adds the plugin configuration to your project's .claude/ directory.
# List installed plugins
/plugin list
# Update to latest version
/plugin update oss-eval
# Disable temporarily
/plugin disable oss-eval
# Re-enable
/plugin enable oss-eval
# Uninstall
/plugin uninstall oss-eval
Run a complete 15-phase evaluation with a single command:
# Full automated evaluation
/oss-eval:run Python web frameworks
# With interactive checkpoints for user decisions
/oss-eval:run JavaScript state management --interactive
# Stop after Phase 6 for interim review
/oss-eval:run container orchestration --to-phase 6
# Resume an interrupted evaluation
/oss-eval:run --resume
For more control, run phases individually:
# Initialize a new evaluation
/oss-eval:start Python web frameworks
# Check current progress
/oss-eval:status
# Work through phases sequentially
/oss-eval:discover # Phase 1
/oss-eval:analyze # Phase 2
/oss-eval:matrix # Phase 3
# ... continue through all 15 phases
# Generate reports
/oss-eval:report --interim # After Phase 6
/oss-eval:report # After Phase 15
| Phase | Command | Purpose |
|---|---|---|
| 1 | /oss-eval:discover | Web search discovery of candidates |
| 2 | /oss-eval:analyze | Detailed candidate analysis |
| 3 | /oss-eval:matrix | Feature comparison matrix |
| 4 | /oss-eval:licensing | License and dependency analysis |
| 5 | /oss-eval:community | Community health assessment |
| 6 | /oss-eval:risk | Quantified risk scoring |
Milestone: After Phase 6, generate interim report with /oss-eval:report --interim
| Phase | Command | Purpose |
|---|---|---|
| 7 | /oss-eval:architecture | Code-level architecture analysis |
| 8 | /oss-eval:requirements | Requirements alignment mapping |
| 9 | /oss-eval:gaps | Gap mitigation strategy |
| Phase | Command | Purpose |
|---|---|---|
| 10 | /oss-eval:ui | UI integration assessment |
| 11 | /oss-eval:costs | Total cost of ownership |
| 12 | /oss-eval:dx | Developer experience evaluation |
| Phase | Command | Purpose |
|---|---|---|
| 13 | /oss-eval:context | Product context integration |
| 14 | /oss-eval:hybrid | Hybrid strategy exploration |
| 15 | /oss-eval:validate | Adversarial review & validation |
Final: Generate complete report with /oss-eval:report
Three core skills loaded automatically during evaluation:
bias-prevention: 12 pitfall prevention rulesfeature-verification: Cross-platform capability checkingsource-validation: Web source credibility verification