MCP App StoreMCP App Store
from prd-debate21
Run a multi-agent, cross-model adversarial debate to produce a high-quality product requirements document (PRD) and decision log.
This skill orchestrates a structured, multi-agent debate to produce product documentation (PRD), summaries, and a decision log. It assigns three roles—Host (orchestrator), Proposer (generates proposals via Codex/GPT-5.4), and Reviewer (Claude-based critic)—and runs iterative proposal/critique rounds, framework design, optional codebase research, and final synthesis into a PRD. The emphasis is always product-level decisions (what and why), not technical implementation.
Use this skill when you need to: create or refine a PRD, make complex product or strategy decisions requiring multiple perspectives, compare competing product options, or run a structured decision process that benefits from adversarial review. Not suitable for simple single-shot docs, code review, or low-level technical architecture design.
Designed for environments with Codex CLI and Claude Code support; expects a proposer model (GPT-5.4/Codex) and a Claude reviewer.
A structured multi-agent debate skill for producing high-quality PRDs via adversarial rounds between a Proposer (GPT-5.4 via Codex CLI) and Reviewer (Claude). No bundled scripts to test. SKILL.md is very detailed with clear phases, state management, and constraints. Uses codex exec --full-auto which auto-executes without confirmation, and variable interpolation in bash command templates presents shell injection risk.
Well-crafted skill with thoughtful structure. Shell injection risk from unquoted variable interpolation in bash command templates. The --full-auto flag on codex exec bypasses human confirmation. Skill is entirely in Chinese which limits global usability.
Maestro Orchestrate
Orchestrates complex engineering workflows with enforced design and approval gates, child-agent delegation, and session state management.
Run Execute
Orchestrates execution of work items across modes (autopilot, confirm, validate) with scripted init/complete tooling, plan/test/report artifacts and strict gati
Caveman Compress
Compress natural-language memory files into a compact 'caveman' format to save LLM tokens while preserving code, links, and structure.
Ip2Location IO Automation (Composio)
Automate Ip2location IO tasks via Composio's Rube MCP toolkit — discover tools, verify connections, and execute schema-compliant workflows safely.
SHIP Score Interview
Run a structured SHIP interview to score and prioritise product features or bugs (Income + Strategy - Perspiration). Guides the agent to reach 90-95% confidence
Codex Collab
Bridge Claude and Codex to run tasks, perform code reviews, and run background jobs for research or implementation workflows.
Smalltalk MCP
Connects an LLM (Claude/OpenAI/etc.) to a live Smalltalk (Squeak/Cuis) image over MCP for browsing, evaluating and modifying Smalltalk code.
Weekly Retro (Personal Corp)
Run a structured weekly retrospective: gather facts from repos and tools, interview the founder, and turn decisions into canonical files and issues.