# snyk/agent-scan

Security scanner for AI agents, MCP servers and agent skills.

Repository: https://github.com/snyk/agent-scan
Canonical: https://ross.abutalabs.com/products/agent-scan
Language: Python
License: Apache-2.0
License Family: permissive
Topics: agent, ai, mcp, modelcontextprotocol, security
Last push: 2026-08-26T14:54:39+00:00

## Health v2 (maintenance only)
Score: 82/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 86, longevity 36
- inputs: {"age_days": 513, "days_push": 7, "days_rel": 14, "gap_med": 3, "n_releases_24m": 66}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2958, forks 260 (observed 2026-08-28T04:07:32.296758+00:00)

## What it is
Snyk Agent Scan is a Python-based CLI security scanner that discovers installed AI agent components (agent harnesses, MCP servers, and agent skills) on a machine and scans them for threats like prompt injections, sensitive data handling, and malware hidden in natural language. It is developed by Snyk and distributed via uvx or standalone binaries.

## Use cases
- scan MCP servers for prompt injection vulnerabilities
- audit installed agent skills for malware payloads
- discover AI agent components on my machine
- check agent harnesses for sensitive data handling risks
- security scan my model context protocol setup
- vet third-party agent skills before installing them

## When to choose
- you install or develop MCP servers or agent skills and want to detect prompt injections and hidden malware
- you want a quick local CLI audit of all agent components on a machine
- you are evaluating AI agent supply-chain risk before adopting tools

## When to avoid
- you need stable CLI output fields or issue codes for production automation, since output is experimental and may change
- you need continuous enterprise-scale agent security management rather than local scanning
- you need a non-Python or npm-distributed tool

## Facets
- artifact type: cli-tool
- maturity: experimental
- function: security, vulnerability-scanning, cli, developer-tools
- domain: security, developer-tools, large-language-models
- platform: cli, python, cross-platform, windows
- tags: mcp, agent-security, prompt-injection, sast, ai-supply-chain, snyk, ai-agents, macos, linux

## Member repositories
- snyk/agent-scan (main) score 82

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:32.296758+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:32:12.054464+00:00, confidence not recorded.
  - readme: https://github.com/snyk/agent-scan (fetched 2026-08-28T04:07:32.296758+00:00, sha dc3b4300a95a)
- Data as of 2026-08-30T08:39:29.467469+00:00.