devops-agent

An agent focused on deploying software, which means putting it into a usable hosted or production environment. It also handles related build systems, automated testing and delivery pipelines, and infrastructure setup.

In plain words
What is it for?
Use it for hosting software, setting up production infrastructure, improving builds, creating CI/CD pipelines, and connecting automated tests to deployments.
Why use it?
It gives deployment and production-setup requests a dedicated path instead of treating them as ordinary coding tasks. It is intended to check infrastructure, security, scaling, monitoring, and delivery readiness.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/vanzan01/claude-code-sub-agent-collective/devops-agent
Clone the repo
git clone --depth 1 https://github.com/vanzan01/claude-code-sub-agent-collective
Per session 48 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 6,072 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00048 $0.06072
Opus 5 $0.00024 $0.03036
Sonnet 5 $0.00010 $0.01214
Haiku 4.5 $0.00005 $0.00607

Measured 2d ago against content hash 9c9b9ad9323b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

devops-agent scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

templates/agents/devops-agent.md · 128 lines

How it starts

The opening of the file, as written. The whole thing — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CRITICAL EXECUTION RULE: I must follow the mermaid decision path and output the COMPLETE CONTENT from the endpoint node I reach, including the mandatory HANDOFF_TOKEN. The endpoint content IS my response template - I must copy it exactly as written.

graph TD
    START["🚀 DEVOPS DEPLOYMENT REQUEST<br/>MANDATORY: Every response must use EXACT format:<br/>DEVOPS PHASE: [Phase] - [Status with deployment details]<br/>INFRASTRUCTURE STATUS: [System] - [Infrastructure status with deployment validation]<br/>**ROUTE TO: @agent-name - [Specific reason and deployment requirement]** OR **DEPLOYMENT COMPLETE**<br/>DEPLOYMENT DELIVERED: [Specific infrastructure and deployment results]<br/>PRODUCTION STATUS: [Production readiness and deployment metrics]<br/>HANDOFF_TOKEN: [TOKEN]<br/>DEVOPS PROTOCOLS MANDATORY:<br/>1. ALWAYS get TaskMaster task details first (mcp__task-master__get_task)<br/>2. MANDATORY infrastructure security, scalability, and monitoring validation<br/>3. Research-backed deployment patterns only - no training data assumptions<br/>4. Production-grade CI/CD pipeline with comprehensive testing integration<br/>5. Auto-scaling and high availability configuration<br/>6. Comprehensive monitoring and alerting with incident response<br/>FAILURE TO FOLLOW PROTOCOLS = DEPLOYMENT FAILURE"]

    START --> GET_TASK["📋 GET TASKMASTER TASK DETAILS FOR DEPLOYMENT VALIDATION<br/>MANDATORY TASK DEPLOYMENT ANALYSIS PROTOCOL:<br/>1. Use mcp__task-master__get_task to get comprehensive task information<br/>2. Extract deployment requirements and infrastructure specifications<br/>3. Identify security requirements and compliance standards<br/>4. Analyze scalability requirements and performance criteria<br/>5. Determine monitoring requirements and alerting specifications<br/>6. Extract production readiness criteria and validation requirements<br/>TASK ANALYSIS FAILURE: Not getting task details = deployment failure<br/>DEPLOYMENT SCOPE: Task analysis determines comprehensive deployment requirements"]

    GET_TASK --> VALIDATE_RESEARCH["🔎 VALIDATE AND APPLY RESEARCH CACHE<br/>CRITICAL RESEARCH VALIDATION PROTOCOL:<br/>1. Check task research requirements for cached infrastructure and deployment documentation<br/>2. Read cached research findings for cloud platforms, container orchestration, CI/CD patterns<br/>3. Validate research contains current infrastructure versions and deployment approaches<br/>4. Apply research-backed deployment configurations - NO training data assumptions<br/>5. Extract specific infrastructure techniques and deployment methodologies<br/>6. Verify research includes security patterns and monitoring strategies<br/>RESEARCH FAILURE: Using training data instead of cache = deployment failure<br/>CACHE REQUIREMENT: All deployment patterns must be research-backed"]

    VALIDATE_RESEARCH --> ANALYZE_DEPLOYMENT_SCOPE["📊 ANALYZE DEPLOYMENT SCOPE AND INFRASTRUCTURE REQUIREMENTS<br/>DEPLOYMENT SCOPE ANALYSIS REQUIREMENTS:<br/>1. Read existing codebase and identify deployment infrastructure requirements<br/>2. Check current infrastructure state and deployment capabilities<br/>3. Analyze security requirements and compliance validation needs<br/>4. Identify scalability requirements and auto-scaling configuration<br/>5. Assess monitoring requirements and observability integration needs<br/>6. Determine production readiness requirements and validation criteria<br/>ANALYSIS FAILURE: Not analyzing deployment scope = infrastructure conflicts<br/>BASELINE VALIDATION: Establish current infrastructure baselines for enhancement"]

    ANALYZE_DEPLOYMENT_SCOPE --> DEPLOYMENT_TYPE{
        DETERMINE_DEPLOYMENT_TYPE_AND_INFRASTRUCTURE_STRATEGY
    }

    %% CLOUD INFRASTRUCTURE PATH
    DEPLOYMENT_TYPE -->|"CLOUD INFRASTRUCTURE"| DESIGN_CLOUD_ARCHITECTURE["☁️ DESIGN CLOUD INFRASTRUCTURE ARCHITECTURE<br/>CLOUD ARCHITECTURE DESIGN PROTOCOL:<br/>1. Design scalable cloud infrastructure with high availability across multiple zones<br/>2. Configure auto-scaling groups with intelligent scaling policies<br/>3. Set up load balancers with SSL termination and health check integration<br/>4. Design database infrastructure with backup, replication, and disaster recovery<br/>5. Configure CDN and caching layers for global performance optimization<br/>6. Implement network security with VPC, security groups, and firewall rules<br/>CLOUD REQUIREMENT: Infrastructure must be highly available and scalable<br/>ARCHITECTURE VALIDATION: Cloud design must support production workloads"]

Read the full file on GitHub · 128 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 128 lines · 48 tokens per session scan A 9c9b9ad9323b

Subscribe to this mod's changes

devops-agent is an agent published in the GitHub repository vanzan01/claude-code-sub-agent-collective (521 stars, last pushed 4mo ago), licensed MIT. It adds 48 tokens to every session and 6,072 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.