AgentSuite — Development skill for Claude Code
Seven role-specific reasoning agents that turn vague intent into precise operating artifacts.
How to install AgentSuite
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open scottconverse/AgentSuite and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What AgentSuite does
Seven role-specific reasoning agents that turn vague intent into precise operating artifacts.
Alternatives in Development
- LifeOS — ⛰️ The Life Operating System — an intent engineering platform that moves you from your current state to your i 18.7k ★
- Ensue Skill — A persistent knowledge tree that grows with you - what you learn today enriches tomorrow's reasoning 411 ★
- Fable OS — An agentic operating system where the kernel is controlled directly by Claude 324 ★
README
AgentSuite
Seven role-specific reasoning agents that turn vague intent into precise operating artifacts.
**v1.1.1** — Specification Kernel + Founder · Design · Product · Engineering · Marketing · Trust/Risk · CIO Agents + multi-agent pipeline orchestration + intra-stage progress events
Which one do I want? AgentSuite or productteam?
These two projects come from the same author and look similar at first glance — both run a multi-agent pipeline with human approval gates. They're built for different jobs.
| What it does | Best for | |
|---|---|---|
| AgentSuite | 7 reasoning agents (Founder, Design, Product, Engineering, Marketing, Trust/Risk, CIO) produce reusable artifacts: brand systems, brief libraries, voice guides, prompt templates, engineering specs. Python package + MCP server. The output is a persistent specification kernel that feeds every future AI session. | Technical founders and developers building inside AI-native IDEs (Claude Code, Cowork, Codex) who are frustrated that every session re-drifts on voice and re-argues settled decisions. You want a kernel that remembers, not one-off generations. |
| productteam | 7-agent software delivery pipeline: concept → PRD → sprint plan → implementation passes → evaluation passes → documentation. The builder never grades its own work; a skeptical evaluator runs tests and tries to break things. Local (Ollama) or cloud (Anthropic / OpenAI / Gemini). Three human gates confirm intent, scope, readiness. | Anyone shipping a small project who wants the agent stack to write the code, not the spec. PRD-to-shipped-and-tested in supervised passes. State persistence + recovery tools for long runs. |
**TL;DR.** AgentSuite produces *specifications and artifacts*. productteam produces *running code*. They compose well: use AgentSuite to define
Related Skills
Yotta Prompt
元引 yotta-prompt — intent clarification & ecosystem entry Agent Skill: turns vague prompts into runnable tasks
Video Editing Suite
Seven Claude Code / Cowork skills that turn a long recording into finished short-form video: transcribe, find
ExportProfile
Generate a complete personal profile summary of everything you know about me from our conversations. Be thorou
Lazyprompt
Codex skill + plugin: turn vague ideas into optimal prompts for lazycodex/OmO $ulw-plan and $ulw-loop
Superforecasting Skill
Turn vague concerns into resolvable, updatable, scoreable probabilistic forecasts. A Claude Code / Codex skill
Fde Simulation
Hands-on Forward Deployed Engineer role simulations: scope an enterprise engagement, build multi-agent prototy
Related Agents
Sonmat Witness
External witness agent. Verifies intent-artifact match using user turn cascade and ground truth. Protocol-isol
Realestate Estimator Sow
Turns vague client briefs and property needs into precise Statements of Work with hours, risk buffers, timelin
Feature Architect
Use to turn a vague client ask into a buildable feature spec — scoping, decisions-needed identification, skill
