dansasser/claude-code-marketplace
Agent
Use when analyzing large files (>20KB), multiple file references, or complex reviews with ollama-prompt. Automatically estimates tokens, chunks if needed, and synthesizes combined analysis.
Dan Sasser's Claude Code plugin marketplace. AI-powered development tools including multi-agent systems, code analysis, and workflow automation. Currently features claude-ollama-agents: intelligent delegation to local ollama models that saves 70%+ context budget through parallel processing and automatic model selection.
dansasser/claude-code-marketplace
Agent
Use when analyzing large files (>20KB), multiple file references, or complex reviews with ollama-prompt. Automatically estimates tokens, chunks if needed, and synthesizes combined analysis.
dansasser/claude-code-marketplace
Agent
Decomposes deep analysis tasks into parallel perspectives (max 4 angles), executes them concurrently with chunking when needed, and manages session continuity for flexible combination strategies.
dansasser/claude-code-marketplace
Agent
Meta-orchestrator that decides whether to use ollama-prompt, which model to select (kimi-k2-thinking, qwen3-vl, deepseek), and whether to delegate to ollama-chunked-analyzer for large tasks. Use when user requests analysis, reviews, or tasks that might benefit from specialized models.
dansasser/claude-code-marketplace
Agent
Use this agent when the user wants to apply CodeRabbit AI review suggestions to a pull request. Triggers when the user says "fix PR", "apply coderabbit fixes", "fix coderabbit review", provides a GitHub PR URL and asks to fix or apply review comments, or any similar phrasing indicating they want CodeRabbit suggestions…
dansasser/claude-code-marketplace
Agent
Use this agent when the user wants to apply OpenAI Codex review suggestions to a pull request. Triggers when the user says "fix codex review", "apply codex fixes", "fix codex PR", provides a GitHub PR URL and mentions codex, or any similar phrasing indicating they want Codex (chatgpt-codex-connector) suggestions…
dansasser/claude-code-marketplace
Agent
Generates detailed PR content with Mermaid diagrams, CI results tables, and changelog entries.
dansasser/claude-code-marketplace
Agent
Main orchestrator for PR preparation. Coordinates all 9 phases (0-8) from unknown branch state to published package. Runs local CI before push, with smart error recovery.
dansasser/claude-code-marketplace
Agent
Gate 6 - API compatibility agent. Detects breaking changes in public API. PREREQUISITE: Gates 1-5 must pass.
dansasser/claude-code-marketplace
Agent
Gate 2 - Test coverage enforcement agent. Ensures test coverage meets threshold. PREREQUISITE: Gate 1 (lint-test) must pass first.
dansasser/claude-code-marketplace
Agent
Gate 3 - Cross-platform compatibility agent. Validates code works on Windows and Ubuntu. Detects path issues, line endings, environment variables. PREREQUISITE: Gates 1-2 must pass.
dansasser/claude-code-marketplace
Agent
Advisory agent for dependency health checks. No prerequisites - can run anytime for dependency analysis and recommendations.
dansasser/claude-code-marketplace
Agent
Gate 8 - GitHub PR agent. Creates PR with full gate report and Mermaid diagrams. PREREQUISITE: ALL gates 1-7 must pass. REFUSES to run otherwise.
dansasser/claude-code-marketplace
Agent
Gate 1 - Linting and testing agent. Runs ruff, mypy, and pytest. Use PROACTIVELY when code quality checks are needed. No prerequisites required.
dansasser/claude-code-marketplace
Agent
Gate 7 - Package build agent. Validates package builds, installs, and entry points work. PREREQUISITE: Gates 1-6 must pass.
dansasser/claude-code-marketplace
Agent
Advisory agent for architecture analysis and tech debt identification. No gate prerequisites - can run anytime for codebase insights.
dansasser/claude-code-marketplace
Agent
Pipeline orchestrator that coordinates all 8 quality gates. Use when running full preflight checks or managing gate sequence. PROACTIVELY invoked for /preflight and /ship commands.
dansasser/claude-code-marketplace
Agent
Gate 4 - Python version matrix agent. Tests against Python 3.9-3.13. PREREQUISITE: Gates 1-3 must pass.
dansasser/claude-code-marketplace
Agent
Gate 5 - Security audit agent. Scans for secrets, vulnerabilities, and license issues. PREREQUISITE: Gates 1-4 must pass.