SPOTTER MCP

Ansible and Terraform governance, built into your AI coding agent

Catch errors, apply fixes, and enforce your policies on IaC code your AI agent writes. Spotter connects to Claude Code, GitHub Copilot, and more.

Spotter MCP in an AI coding agent: the agent writes an Ansible Playbook, runs spotter_scan, and resolves all errors and warnings
  • Checks inside the agent’s workflow
    The agent checks everything it writes, including unsaved content.
  • Automatic fixes and rechecks
    Spotter fixes what it can automatically. The agent handles the remaining findings and runs another check.
  • Your organization’s policies
    Define your rules once, and your agent follows them in every scan: custom policies, required checks and paths to skip.
  • Built-in guardrails
    Agents can’t skip a check to make a result pass or edit content outside the task.
TRUSTED BY
KEY BENEFITS

Ship AI-written automation you can trust

Review finished work, not the agent's mistakes
Your agent hands back content that has already been scanned and fixed, so reviewers spend their time on logic instead of catching basic errors.
Keep AI within your rules
Every playbook your agent writes follows your organization’s security, compliance and quality policies, without depending on how a prompt was phrased.
Catch issues before production
Deprecated modules, version incompatibilities and risky patterns are found and fixed while the code is being written, not after deployment.
HOW IT WORKS
Check, fix and recheck before it reaches production
AI agents don't automatically know your organization's rules. Spotter brings your existing policies into their workflow, checking and fixing code before you review it.
Check, fix and recheck
Your policies
YOUR RULES
Your policies, applied to agent output
The agent reads your project’s Spotter config and applies the rules you’ve already defined, with nothing new to configure or maintain.
BUILT FOR AGENTS
Designed for how agents work
Spotter returns concise, prioritized results that the agent can work through right away, keeping every run fast and light on tokens.
  • Findings grouped by check and ordered by severity
  • Fixable checks clearly identified
  • Results limited to 40 lines
Built for agents

Get started

Bring Spotter into your AI agent. Request access and get up and running in one install.

Get access

Contact our team

Get all questions answered and learn about Steampunk Spotter for enterprises.

  • Schedule a demo
  • Get pricing information
  • Explore use cases for your team
Beams
Beams

Something went wrong. Please try again later.

Please wait

Processing, please wait...

Thank you for reaching out to us

We will get back to you shortly.

FAQ

A Python package (steampunk-spotter-mcp on PyPI) that gives any MCP-compatible AI agent one tool, spotter_scan. The agent uses it to check Ansible and Terraform content with Spotter, apply automatic fixes and confirm the result.
Spotter MCP is available on request. Fill in the access form and our team will get you set up.
The MCP server gives the agent the tool. The agent plugin (spotter) bundles that server with the spotter-scan skill, which tells the agent when and how to use it. Without the skill, agents often don’t check at all, so we recommend the plugin.
Yes. The agent checks Terraform configurations as well as Ansible Playbooks, with the same Spotter account and organization policies.
Claude Code and GitHub Copilot, plus any agent that supports the Model Context Protocol (MCP).
No. Both use your existing Steampunk Spotter account, plan and policies, the same as the CLI.
Yes. It works with on-prem installations as well as Spotter in the cloud.