# Send DevOps to your agent
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
## Fast path
- Download the package from Yavira.
- Extract it into a folder your agent can access.
- Paste one of the prompts below and point your agent at the extracted folder.
## Suggested prompts
### New install

```text
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete.
```
### Upgrade existing

```text
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run.
```
## Machine-readable fields
```json
{
  "schemaVersion": "1.0",
  "item": {
    "slug": "devops",
    "name": "DevOps",
    "source": "tencent",
    "type": "skill",
    "category": "开发工具",
    "sourceUrl": "https://clawhub.ai/ivangdavila/devops",
    "canonicalUrl": "https://clawhub.ai/ivangdavila/devops",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadUrl": "/downloads/devops",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=devops",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "packageFormat": "ZIP package",
    "primaryDoc": "SKILL.md",
    "includedAssets": [
      "SKILL.md"
    ],
    "downloadMode": "redirect",
    "sourceHealth": {
      "source": "tencent",
      "slug": "devops",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-05-02T11:58:34.583Z",
      "expiresAt": "2026-05-09T11:58:34.583Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=devops",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=devops",
        "contentDisposition": "attachment; filename=\"devops-1.0.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null,
        "slug": "devops"
      },
      "scope": "item",
      "summary": "Item download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this item.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/devops"
    },
    "validation": {
      "installChecklist": [
        "Use the Yavira download entry.",
        "Review SKILL.md after the package is downloaded.",
        "Confirm the extracted package contains the expected setup assets."
      ],
      "postInstallChecks": [
        "Confirm the extracted package includes the expected docs or setup files.",
        "Validate the skill or prompts are available in your target agent workspace.",
        "Capture any manual follow-up steps the agent could not complete."
      ]
    }
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/devops",
    "downloadUrl": "https://openagent3.xyz/downloads/devops",
    "agentUrl": "https://openagent3.xyz/skills/devops/agent",
    "manifestUrl": "https://openagent3.xyz/skills/devops/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/devops/agent.md"
  }
}
```
## Documentation

### CI/CD Pipelines

Fail fast: run linting and unit tests before expensive integration tests — saves time and compute
Cache dependencies between runs — npm install on every build wastes minutes
Pin action versions with SHA, not tags — actions/checkout@v3 can change, SHA is immutable
Secrets in environment variables, never in code or logs — mask them in CI output
Parallel jobs for independent steps — test, lint, and build can run simultaneously

### Deployment Strategies

Blue-green: run new version alongside old, switch traffic atomically — instant rollback by switching back
Canary: route percentage of traffic to new version — catch issues before full rollout
Rolling: update instances incrementally — balance between speed and risk
Always have rollback plan before deploying — know exactly how to revert
Deploy the same artifact to all environments — build once, promote through stages

### Infrastructure as Code

Version control all infrastructure — terraform, ansible, cloudformation in git
Never apply changes without plan/diff review — terraform plan before apply
State files contain secrets — store remotely with encryption, never in git
Modules for reusable components — don't copy-paste infrastructure definitions
Separate environments with workspaces or directories — dev changes shouldn't affect prod

### Containers

One process per container — containers are not VMs
Health checks are mandatory — orchestrators need them for routing and restarts
Don't run as root — use non-root USER in Dockerfile
Immutable images: config via environment, not baked in — same image in all environments
Tag images with git SHA, not just latest — know exactly what's deployed

### Secrets Management

Never store secrets in environment files committed to git — use vault, sealed secrets, or CI secret storage
Rotate secrets regularly — automation makes rotation painless
Different secrets per environment — dev leak shouldn't compromise prod
Audit secret access — know who accessed what and when
Secrets in memory, not disk when possible — temp files persist longer than expected

### Monitoring & Alerting

Four golden signals: latency, traffic, errors, saturation — start here
Alert on symptoms, not causes — "users seeing errors" not "CPU high"
Every alert must be actionable — if you can't do anything, it's noise
Dashboard per service with key metrics — one glance shows health
Structured logs (JSON) for machine parsing — grep works, but queries are better

### Reliability

Define SLOs before building alerting — what does "healthy" mean for this service?
Error budgets: some failures are acceptable — 99.9% means 8 hours downtime/year is OK
Chaos engineering in staging — break things intentionally before prod breaks accidentally
Runbooks for common incidents — 3am is not the time to figure out recovery steps
Post-mortems without blame — focus on systems, not people

### Common Mistakes

SSH into prod to fix things — all changes through automation, or you'll forget what you did
No staging environment — "works on my machine" doesn't mean works in prod
Ignoring flaky tests — they erode trust in CI, either fix or delete
Manual steps in deployment — if it's not automated, it'll be done wrong eventually
Monitoring only happy paths — check error rates and edge cases too

### Networking

Internal services don't need public IPs — use private subnets, expose only load balancers
TLS everywhere, including internal traffic — zero trust, even behind firewall
DNS for service discovery — hardcoded IPs break when things move
Load balancer health checks separate from app health — LB needs fast response, app health can be thorough
Firewall default deny — explicitly allow what's needed, block everything else
## Trust
- Source: tencent
- Verification: Indexed source record
- Publisher: ivangdavila
- Version: 1.0.0
## Source health
- Status: healthy
- Item download looks usable.
- Yavira can redirect you to the upstream package for this item.
- Health scope: item
- Reason: direct_download_ok
- Checked at: 2026-05-02T11:58:34.583Z
- Expires at: 2026-05-09T11:58:34.583Z
- Recommended action: Download for OpenClaw
## Links
- [Detail page](https://openagent3.xyz/skills/devops)
- [Send to Agent page](https://openagent3.xyz/skills/devops/agent)
- [JSON manifest](https://openagent3.xyz/skills/devops/agent.json)
- [Markdown brief](https://openagent3.xyz/skills/devops/agent.md)
- [Download page](https://openagent3.xyz/downloads/devops)