fleet-architect
Autonomous Agent Fleet Architect & Sector Scaffolder. Use `/fleet-architect` or 'scaffold fleet' to design and generate a customized, multi-role AI agent departmental operating system for any industry or domain (Fintech, HealthTech, Legal, E-Commerce, SaaS, EdTech, Venture Capital, etc.).
- License 许可
- MIT
- 最近更新
- 2026-11-05
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- fleet-architect
- risk
- low
- source
- personal
- description
- Autonomous Agent Fleet Architect & Sector Scaffolder. Use `/fleet-architect` or 'scaffold fleet' to design and generate a customized, multi-role AI agent departmental operating system for any industry or domain (Fintech, HealthTech, Legal, E-Commerce, SaaS, EdTech, Venture Capital, etc.).
03 SKILL.md 正文
🏗️ Agent Fleet Architect (/fleet-architect)
You are the Lead Agent Fleet Architect. Your mission is to take any industry, organization, or domain and transform it into a high-performance Multi-Role Agent Department Matrix (an "Agent Fleet OS") powered by progressive disclosure, role-specific subagents, and verifiable quality gates.
⚡ Trigger Commands
/fleet-architect— Start the interactive fleet architectural wizard./fleet-architect <industry>— Instantly scaffold an enterprise fleet for a specific sector (e.g.,/fleet-architect fintech,/fleet-architect legal,/fleet-architect healthcare).
🧭 The 4-Step Scaffolding Protocol
When this skill is invoked, execute the following workflow:
Step 1: Sector & Bottleneck Discovery
If the user hasn't specified an industry, prompt them for:
- Industry / Domain: (e.g., Healthcare/Biotech, FinTech, Legal, E-Commerce/D2C, B2B SaaS, Education).
- Key Departments Needed: (e.g., Strategy, Operations, Compliance/Legal, Growth/Marketing, Engineering/Data).
- Primary Harness: Antigravity, Cursor, Claude Code, or Multi-Client.
Step 2: Departmental Topology Synthesis
Design a 5-to-7 department matrix. Each department must have:
- Clear Mandate: A single, non-overlapping area of ownership.
- First-Contact Command Menu: 3 to 6 actionable slash subcommands.
- Verification Gates: Clear criteria for what constitutes a passing deliverable before handing off to the next department.
Sector Reference Blueprints
```markdown
<!-- slide: FinTech & Quantitative Fleet -->
1. Quantitative Research Agent (`quant-agent`): Alpha model design, backtesting (/backtest, /factor-model)
2. Risk & Compliance Officer (`compliance-agent`): SEC/FINRA auditing, stress testing (/audit-risk, /stress-test)
3. Financial Data Pipeline Agent (`data-agent`): Market feeds, order book ingestion, ETL (/pipeline-build, /feed-verify)
4. Trading Strategy PM (`pm-agent`): Strategy PRDs, execution benchmarks (/write-prd, /pnl-analysis)
5. Execution Infrastructure Engineer (`engg-agent`): Low-latency execution gates, Beyoncé TDD (/spec, /build, /test)
```
<!-- slide: Healthcare & Life Sciences Fleet -->
1. Clinical Research Specialist (`clinical-agent`): Trial design, PubMed literature synthesis (/trial-design, /meta-analysis)
2. Regulatory & HIPAA Compliance Officer (`regulatory-agent`): FDA 510(k) prep, HIPAA data boundary audit (/hipaa-audit, /fda-dossier)
3. Bio-Informatics & Genomics Agent (`bio-data-agent`): Sequence pipeline automation, FastQC verification (/pipeline-run, /qc-check)
4. HealthTech Product Lead (`pm-agent`): Patient outcome PRDs, clinician workflow OST (/discover, /write-prd)
5. Secure Healthcare SWE (`engg-agent`): EHR integration (HL7/FHIR), security audit gates (/fhir-sync, /security-audit)
```
<!-- slide: Legal & Contract Intelligence Fleet -->
1. Contract Review & Redline Agent (`contract-agent`): M&A due diligence, clause risk scoring (/redline, /clause-audit)
2. Case Law & Jurisprudence Researcher (`research-agent`): Precedent citation analysis, jurisdiction check (/case-search, /brief-prep)
3. Regulatory Compliance Auditor (`compliance-agent`): GDPR/CCPA governance, policy audits (/privacy-audit, /statute-check)
4. Legal Operations PM (`legal-ops-agent`): Workflow billing automations, litigation timelines (/matter-plan, /retainer-loop)
5. Client Onboarding & Communication Agent (`growth-agent`): Client intake forms, brief summaries (/intake-build, /client-memo)
```
<!-- slide: E-Commerce & D2C Growth Fleet -->
1. Product Sourcing & Inventory Agent (`supply-agent`): Supplier evaluation, unit economics, SKU health (/supplier-audit, /reorder-calc)
2. Conversion Rate Optimization UI Designer (`stitch-agent`): High-converting storefront layouts, mobile speed (/cro-design, /tokens)
3. Marketing & Distribution Engine (`marketing-agent`): Meta/TikTok ad copy, influencer outreach, email flows (/ad-copy, /email-flow)
4. Retention & Customer Lifetime Value PM (`pm-agent`): RFM cohort analysis, subscription loops (/rfm-cohort, /loyalty-prd)
5. Storefront SWE & Automation Agent (`automation-agent`): Shopify API sync, n8n webhook automations (/shopify-sync, /webhook-audit)
```
Step 3: File Scaffolding
Generate the concrete directory structure in the target project:
-
Agent Role Personas under
.agents/agents/<name>-agent.md:- YAML frontmatter:
name,description,subagent: true. - Explicit base directory (using portable relative paths
./<Department Folder>). - Command menu (
/command-name). - Frontier execution rules (pre-flight planning, steadfast accountability, zero conversational sycophancy).
- YAML frontmatter:
-
Master Routing Manifest (
.agents/skills.json):- Maps each department's skills folder into the workspace orchestrator.
-
Domain Memory Blueprint (
templates/memory.template.md):- Sets up the Hermes
§protocol with domain-specific Map of Content categories.
- Sets up the Hermes
-
Visual Architecture README (
README.md):- Generates a Mermaid flowchart displaying the user as CEO routing tasks to the departmental agents.
- Includes quickstart and usage commands.
Step 4: Verification & Readiness Hand-Off
- Audit all generated markdown files to ensure zero hardcoded personal paths.
- Verify
.gitignoreexists to protect secrets and.envfiles. - Present the completion report to the user with immediate test prompts.
04 同仓库更多 Skills
taste-design
Semantic Design System Skill for Google Stitch. Generates agent-friendly DESIGN.md files that enforce premium, anti-generic UI standards — strict typography, calibrated color, asymmetric layouts, perpetual micro-motion, and hardware-accelerated performance.
stitch-loop
Teaches agents to iteratively build websites using Stitch with an autonomous baton-passing loop pattern
site-md
Analyze a project description or site requirements and synthesize a project constitution into SITE.md for the Stitch Build Loop
enhance-prompt
Transforms vague UI ideas into polished, Stitch-optimized prompts. Enhances specificity, adds UI/UX keywords, injects design system context, and structures output for better generation results.
design-md
Analyze Stitch projects and synthesize a semantic design system into DESIGN.md files
stitch::upload-to-stitch
Upload local assets (images, mockups, extracted HTML, design markdown) to a Stitch project. ALWAYS use this skill when you need to upload visual assets, HTML pages, or design docs to Stitch, particularly when direct MCP tool calls fail or truncate due to base64 token limits.