deploy-agent-production
Skill0 starsChanged 6 months agoMITView on GitHub
Deploys AI agents and MCP servers to production: Docker, serverless, streamable HTTP MCP, secrets, and health checks. Use when shipping agents to prod, hosting MCP remotely, or DevOps for agentic systems.
- Installs packages
In this repository14 other files it gives its agents.
- a2a-agent-delegationSkill
- agent-memory-patternsSkill
- agent-observabilitySkill
- agent-redteam-securitySkill
- browser-agent-automationSkill
- build-mcp-serverSkill
- create-production-skillSkill
- cursor-sdk-automationSkill
- evaluate-agent-qualitySkill
- github-ci-agentsSkill
- langgraph-multi-agentSkill
- multi-agent-supervisorSkill
- rag-agent-pipelineSkill
- secure-mcp-serverSkill
The file
Files
Deploy Agent Production
Deployment targets
| Component | Common target |
| MCP server (stdio) | Dev only / local IDE |
| MCP server (HTTP) | Container + HTTPS reverse proxy |
| LangGraph API | LangGraph Platform or custom FastAPI |
| Worker agents | Queue + autoscaled workers |
Workflow
- [ ] Containerize with pinned dependencies
- [ ] Secrets via vault / platform secrets (never in image)
- [ ] Health check: GET /health or MCP ping tool
- [ ] Timeouts on all outbound tool calls
- [ ] Autoscaling on queue depth or request rate
- [ ] Rollback strategy documentedMCP remote (streamable HTTP)
- TLS termination at gateway
- OAuth for multi-tenant
- Rate limit per API key
- CORS locked to known clients if browser-accessible
Docker sketch
FROM python:3.12-slim
WORKDIR /app
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt
COPY . .
CMD ["python", "-m", "mcp_server"]
HEALTHCHECK CMD curl -f http://localhost:8080/health || exit 1Production checklist
- Resource limits (CPU/memory)
- Structured logs + trace IDs
- Graceful shutdown (drain in-flight)
- Staging environment mirrors prod tools (sandbox APIs)
See secure-mcp-server before exposing HTTP MCP publicly.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.