Multi-agent SDLC framework with Cognitive Council deliberation and task decomposition patterns for structured feature development
Quick single-pass task decomposition without full council deliberation — faster and cheaper for straightforward features
Decompose an epic/feature into SDLC tasks using a 5-expert cognitive council
Orchestrates a Cognitive Council deliberation for SDLC task decomposition. Use when decomposing an epic, feature, or complex initiative into structured tasks via multi-expert deliberation. Manages 6 domain experts (including an adversarial Red Team agent) through a 5-phase protocol (Problem Framing, Independent Hypothesis, Cross-Critique, Revision, Vote, Synthesis) and produces a COUNCIL_VERDICT with a unified task manifest.
SDLC DevOps/Delivery expert for Cognitive Council deliberation. Analyzes epics from the perspective of CI/CD pipeline, infrastructure, deployment strategy, release sequencing, and feature flags. Use as part of the council-orchestrator's 5-expert deliberation team for task decomposition.
SDLC Implementation Engineer expert for Cognitive Council deliberation. Analyzes epics from the perspective of code-level task granularity, effort estimation, practical sequencing, and developer experience. Use as part of the council-orchestrator's 5-expert deliberation team for task decomposition.
SDLC Product/Design expert for Cognitive Council deliberation. Analyzes epics from the perspective of user journeys, use cases, story mapping, value delivery sequencing, and acceptance from the user's point of view. Use as part of the council-orchestrator's 5-expert deliberation team for task decomposition.
SDLC QA Strategist expert for Cognitive Council deliberation. Analyzes epics from the perspective of test coverage, acceptance criteria, edge cases, quality gates, and regression risk. Use as part of the council-orchestrator's 5-expert deliberation team for task decomposition.
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Private Claude Code plugin marketplace. Hosts the Catalyst family of multi-agent SDLC and workflow plugins.
catalyst-agentic-sdlc/ # repo root = marketplace root
├── .claude-plugin/
│ └── marketplace.json # marketplace catalog (name: "catalyst")
└── plugins/
├── catalyst-sdlc-framework/ # plugin: Cognitive Council SDLC framework
│ ├── .claude-plugin/plugin.json
│ ├── agents/ # 7 council experts + structured-logging-architect
│ └── commands/ # /sdlc-council, /sdlc-council-quick
├── ticket-workflows/ # plugin: structured ticket workflow
│ ├── .claude-plugin/plugin.json # ships PostToolUse/PreToolUse/Stop hooks
│ ├── agents/ # dev, review, qa, coverage
│ ├── commands/ # begin, status, review, qa, tech-compare, wrapup, dev
│ └── templates/ # context, pr-description, qa-report
└── catalyst-meta/ # plugin: global commands, hooks, MCP servers
├── .claude-plugin/plugin.json # wires SessionStart/PreCompact/PreToolUse hooks
├── .mcp.json # bundled MCP server stack
├── commands/ # memory-status, init-memory, save-context, optimize
├── hooks/ # workstream-briefing, pre/post-compact, git-guard
└── docs/ # memory-architecture reference
The repo follows the multi-plugin marketplace pattern from the Claude Code plugin marketplace docs: a single marketplace.json at the root lists every plugin, each plugin lives under plugins/<name>/ with its own .claude-plugin/plugin.json and component directories (agents/, commands/, skills/, hooks/).
# During development (local path)
/plugin marketplace add /Users/panda/catalyst-devspace/workspace/catalyst-sdlc-framework
# From GitHub
/plugin marketplace add TheBranchDriftCatalyst/catalyst-agentic-sdlc
/plugin install catalyst-sdlc-framework@catalyst
/plugin install ticket-workflows@catalyst
/plugin install catalyst-meta@catalyst
Installed plugins are cached at ~/.claude/plugins/cache/ — re-run /plugin marketplace update catalyst after structural changes to pull updated manifests.
catalyst-sdlc-frameworkA 6-expert Cognitive Council that decomposes epics and features into structured, sequenced SDLC task manifests through a strict 5-phase deliberation protocol. The goal is to surface dissent, prevent groupthink, and produce a manifest a delivery team can turn into tickets without further refinement.
| Command | Purpose |
|---|---|
/sdlc-council <epic> | Full council deliberation — all 5 phases, all 6 experts. Use for complex initiatives where consensus quality matters. |
/sdlc-council-quick <epic> | Single-pass decomposition for straightforward features. Faster and cheaper, no cross-critique. |
| Agent | Perspective |
|---|---|
sdlc-systems-architect | Component boundaries, API contracts, data models, integration points |
sdlc-impl-engineer | Code-level granularity, effort, sequencing, developer experience |
sdlc-qa-strategist | Test coverage, acceptance criteria, edge cases, quality gates |
sdlc-devops-delivery | CI/CD, infrastructure, release strategy, feature flags |
sdlc-product-design | User journeys, story mapping, value delivery sequencing |
sdlc-red-team | Adversarial — challenges necessity, finds scope creep, argues for minimum viable scope |
| Agent | Purpose |
|---|---|
structured-logging-architect | Senior observability engineer — designs structured logging for Elasticsearch/Datadog. Invoke directly when needed. |
npx claudepluginhub thebranchdriftcatalyst/catalyst-agentic-sdlc --plugin catalyst-sdlc-frameworkCatalyst meta-plugin — global agents, slash commands, and session hooks (workstream briefing, pre/post-compact context preservation, git-guard). Designed to be kept always-enabled.
Structured ticket workflow with progressive context, beads integration, quality gates, and per-stage agents (dev, review, QA, coverage)
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
Comprehensive PR review agents specializing in comments, tests, error handling, type design, code quality, and code simplification
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Complete developer toolkit for Claude Code