Sandboxes and scanners

Pipelock vs Snyk Agent Scan

A pre-deploy scanner that finds installed agents, MCP servers, and skills and checks them for injection and secrets, next to a runtime firewall that inspects the traffic they produce.

At a glance

Pipelock source Snyk Agent Scan
Job Agent firewall. Mediates HTTP, WebSocket, and MCP traffic routed through it, scans it for secret leaks, prompt injection, SSRF, and tool poisoning, and can emit signed action receipts for mediated decisions when a signing key is configured. Discovers the agent components installed on a machine (harnesses, MCP servers, skills) and scans them for prompt injection, sensitive data handling, and malware patterns before they run.
Enforcement point Network path, outside the agent process Pre-deploy, on the developer machine or in CI; it starts MCP servers to read their tool descriptions
Source Open source, Apache-2.0 core; Enterprise under ELv2 Open source, Apache-2.0 (Snyk); the former Invariant Labs mcp-scan repository redirects here
Pricing shape Free core; paid Pro and Enterprise tiers Free scanner; runs with a Snyk API token
Runs as Single Go binary, self-hosted; container and Helm uvx or release binary; PyPI package
Pick Snyk Agent Scan

You want an inventory of every agent, MCP server, and skill on a machine, scored for injection, secret handling, and malware, before anything is trusted.

Pick Pipelock

You need the traffic those components produce at runtime inspected on a routed path, and signed receipts when configured.

Run both

Scan before you trust, inspect while it runs. Agent Scan catches a poisoned skill at install time; Pipelock catches the drift that appears after the scan.

Want the runtime boundary, not just another checklist?

The short version

Pipelock is an open-source agent firewall. It inspects HTTP, WebSocket, and MCP traffic routed through it at runtime for credential leaks, prompt injection, SSRF, and tool poisoning, and can emit signed action receipts for mediated decisions when a signing key is configured.

Snyk Agent Scan is an Apache-2.0 scanner from Snyk. Its README describes discovering the agent components installed on a machine, including harnesses, MCP servers, and skills, and scanning them for prompt injections, sensitive data handling, and malware patterns. It installs with uvx or a release binary, runs with a Snyk API token, and warns that scanning an MCP configuration executes the commands defined in it, because that is how it reads the tool descriptions. The Invariant Labs mcp-scan repository redirects here.

Agent Scan checks what is installed before you trust it. Pipelock checks what it does while it runs. Snyk’s announcement also names Agent Guard, a runtime policy enforcer in private preview; this page compares the open-source scanner only, because the guard is in private preview and its behavior is not documented publicly.

Feature comparison

FeaturePipelockSnyk Agent Scan
When it runsContinuously, on live trafficPoint in time, on demand or in CI
What it examinesHTTP, WebSocket, and MCP payloads as they flowInstalled harnesses, MCP servers, and skills
Tool-description poisoningScanned at runtime, fingerprinted, drift flaggedScanned at scan time
Rug-pull drift after installDetected when a description changes mid-sessionNot a runtime function; re-scan to see it
Credential scanning (DLP)65 patterns, encoding-aware, on live trafficSensitive data handling checks on components
Prompt injectionDeterministic patterns with multi-pass normalization on responsesInjection checks on component content
Malware patterns in skillsNot in scopeYes
SSRF protectionYesNot in scope
EnforcementBlocks or warns on mediated trafficReports; not on the traffic path
Signed receiptsYes, Ed25519, verifiable offlineReports
LicenseApache-2.0 core; Enterprise under ELv2Apache-2.0

Where Snyk Agent Scan is stronger

Discovery. It finds what is installed, across harnesses, servers, and skills, which a network proxy never enumerates.

Skill and malware checks. Agent skills are scanned for malware patterns and unsafe handling before they are used. Pipelock doesn’t analyze installed components.

Fits CI. A scan is a report you can gate a pipeline on.

Where Pipelock is stronger

Runtime truth. A component that passed a scan can still change later. Pipelock reads the tool description every session, fingerprints it, and flags drift.

Traffic inspection. Credentials in a request body, an injected web page, an SSRF attempt: none of these are in a component scan. Pipelock inspects the traffic itself.

Enforcement and evidence. Pipelock blocks on mediated paths and can emit signed receipts for mediated decisions when a signing key is configured. A scanner reports.

Scan, then inspect

The two tools stop failures at different moments. A poisoned skill in a marketplace is caught at install time by a scanner. A tool description that changes after the agent connected, or a response that carries an injection, is caught on the wire by a firewall. Teams that care about both run both.

Further reading

Sources checked

Third-party descriptions on this page come from the public materials below, read on the dates shown. Features and pricing change; check the current documentation before you decide.

  • snyk/agent-scan README checked 2026-09-01 · Apache-2.0; discovers harnesses, MCP servers, and skills; scans for prompt injections, sensitive data handling, malware patterns; warns that scanning an MCP configuration executes its commands; installs via uvx or release binaries
  • snyk/agent-scan releases checked 2026-09-01 · v0.6.1 from August 2026 was the newest release at the time of checking
  • Snyk blog: introducing Agent Security checked 2026-09-01 · Agent Scan in open preview; Agent Guard, a runtime policy enforcer, in private preview; Agent Red Teaming in open preview

Third-party product names and marks belong to their owners. PipeLab is not affiliated with, sponsored by, or endorsed by the makers of any product compared on this page. Descriptions of other products come from their own public materials on the dates listed above and reflect PipeLab's reading of them. If something here is wrong or out of date, tell us and it will be corrected.

Frequently asked questions

What's the difference between Pipelock and Snyk Agent Scan?
Agent Scan is a scanner. It discovers the agents, MCP servers, and skills installed on a machine and checks them for prompt injection, sensitive data handling, and malware patterns before they are used. Pipelock is a runtime agent firewall. It inspects HTTP, WebSocket, and MCP traffic routed through it as the agent works and can emit signed receipts when a signing key is configured. One runs before deployment, the other during.
Does Agent Scan run at runtime?
The open-source scanner is a point-in-time scan. To read tool descriptions it starts the MCP servers defined in a configuration, which its README flags as executing those commands. It does not sit on the network path or mediate live traffic.
Can I use both?
Yes, and the combination is the natural one. Scan components before trusting them, then inspect the traffic they produce. A tool description that changes after the scan is exactly the case a runtime firewall exists for.

Want the runtime boundary, not just another checklist?

See all comparisons →