ZipLoom
Home
PricingSign InGet Your Live URL

own

MCP Server For Coding Agents

Ask Your Coding Agent Whether It's Safe To Ship — And Get A Real Answer.

Updated

At a glance

One MCP endpoint for every major coding agent and editor: scan a project, read findings and deploy status, and call read-only GitHub tools scoped to your repos.

  • The Scanner, Inside The Editor
  • One Endpoint, Every Major Agent
  • Read-Only GitHub Tools, Fenced To Your Repos
Sample. Ask your coding agent whether it's safe to ship. Over MCP it lists your projects, scans one and reads the findings, deploy status and live posture, so the answer comes from a real scan.

Why it matters

Usually nothing but the agent that wrote the code looks at it before you merge. It will say the change looks fine, because it cannot see what a scanner sees.

Connect your coding agent to ZipLoom over MCP and it can scan the commit it just wrote, wait for the result, and read the findings with file, line and fix, without you leaving the editor. It can also read attack chains, deploy status with each gate's verdict and the live posture of the deployed URL. Agent scans reuse a scan already running and are capped per hour, so a looping agent cannot burn your quota.

Your Agent Checks Its Own WorkIllustration · sample data
editoragentwrote a changeMCPscan → findingsfile · line · fixchains · gates · posturechange? needs approvalscans capped per hour

What you get

The Scanner, Inside The Editor

Your agent can start a ZipLoom scan of the commit it just wrote, wait for it, and read the findings — title, severity, file, line and fix — without you leaving the editor or pasting anything.

One Endpoint, Every Major Agent

The leading coding agents and editors connect natively over Streamable HTTP; anything else connects through mcp-remote. Copy-paste setup for each is on the MCP page and in Settings → API Keys.

Read-Only GitHub Tools, Fenced To Your Repos

35 read-only tools from the Agent Ops registry — files, pull requests, workflow runs, security advisories — run under the key creator's GitHub access and only against repositories your organization has connected. 21 more are request-only: they file a change for an owner or admin to approve in ZipLoom, and nothing over MCP changes GitHub by itself. Sensitive changes (merging, repository settings, access, secrets, webhooks, workflow runs, anything under .github/) are dashboard-only.

How It Connects

  1. 1

    Create an IDE Agent key in Settings → API Keys (Owner or Admin). It is shown once and stored hashed.

  2. 2

    Paste the setup snippet for your agent — it reads the key from an environment variable or a secret prompt, never from the file.

  3. 3

    Your agent lists your projects, scans one, and reads findings, attack chains, deploy status and live posture.

  4. 4

    Agent scans reuse an in-flight scan and are capped per hour, so a looping agent can't burn your CI quota.

What this check inspects

  • The same code, secrets, dependency and workflow scan your deploys and pull requests get — one source of truth, three entry points.

  • Deploy status with each gate's verdict — passed, blocked, advisory or error — so an agent never reads a failed-open gate as a pass.

  • Live posture of the deployed URL, so the agent can check the running app, not just the code.

What it means when this fails

Without it, the agent that wrote the code is the only reviewer it gets before you merge. It will tell you the change looks fine, because it cannot see what a scanner sees — and the first independent check happens after the code is already live.

What it does not do

  • The agent cannot change your repository by itself. It can file a request for a routine change, and nothing happens on GitHub until an Owner or Admin approves it.

  • Deploys cannot be started over MCP, and destructive or sensitive tools are not listed.

  • The agent never sees how detection works, only what is wrong, where and how to fix it.

Questions

Can the agent change my repository or deploy through MCP?
Not by itself. Over MCP the agent can read, start a scan, and file a request for a routine, non-destructive change on a connected repository; nothing happens on GitHub until an owner or admin approves the request in ZipLoom. Deploys can't be started over MCP, and destructive and sensitive tools (deletes, removals, merging, repository settings, access, secrets, workflow runs, changes under .github/) are not listed and are refused if called.
Is a CI key enough?
No — CI keys are scan-only and deliberately cannot reach the MCP endpoint. Create an IDE Agent key for agents.
Does the agent see how detection works?
No. It gets what is wrong, where and how to fix it. Detection internals never enter an agent's context, so they can't be talked out of it.

Related features