Automate vulnerability scanning, SBOM generation, and CIS benchmarking for MCP agents.
npx skills add https://github.com/msaad00/agent-bom --skill agent-bom
Please help me install this Agent Skill. Skill: agent-bom Source: https://github.com/msaad00/agent-bom/tree/main/integrations/openclaw Command: npx skills add https://github.com/msaad00/agent-bom --skill agent-bom
Open security platform for agentic infrastructure — broad scanning, blast radius, runtime, and trust across MCP discovery, CVEs, SBOMs, CIS benchmarks (AWS, Azure, GCP, Snowflake), OWASP/NIST/MITRE compliance, AISVS v1.0, MAESTRO layer tagging, and vector database security checks. Use when the user mentions vulnerability scanning, MCP server trust, compliance, SBOM generation, CIS benchmarks, blast radius, or AI supply chain risk.
Install via pipx and run agent-bom agents to discover and scan locally.
High-intent search queries and answers about installing and using this skill.
Automated vulnerability scanning and SBOM generation for MCP servers can be performed locally to map blast radius and trust scores. This process enforces local-first data handling and requires explicit user confirmation before executing any optional cloud checks.
Checking AI agents against CIS benchmarks and AISVS v1.0 involves mapping compliance workflows across local environments. You can run CIS benchmarking for AWS, Azure, GCP, and Snowflake to ensure OWASP, NIST AI RMF, and MITRE ATLAS framework mappings.
Yes, you can discover agents across 22+ tools and assess AI supply chain risk using a local-first approach. The system redacts environment variables from config data and enforces Sigstore provenance verification without requiring cloud connectivity.
CVE enrichment and trust assessment work by automating end-to-end vulnerability scanning across discovered MCP-powered AI agents. The system calculates blast radius and trust scores to evaluate risk, applying MAESTRO layer tagging for comprehensive infrastructure analysis.
Cloud credentials are optional for running CIS compliance checks across AWS, Azure, GCP, and Snowflake. The system requires explicit user confirmation before performing any cloud-based checks, maintaining strict local-first data handling for security.
Environment variables are redacted during MCP discovery to enforce local-first data handling and prevent sensitive configuration data leakage. This security measure ensures safe trust assessment and blast radius mapping without exposing underlying infrastructure credentials.
nadicodeaiDiscover MCP servers and invoke their tools from the command line.
NVIDIAGenerate CycloneDX SBOMs with Syft, resolve licenses, and export CSV reports.
marquesfelipAutomate artifact signing and provenance verification with Sigstore and Cosign.
statick88Automate vulnerability detection across code, dependencies, containers, and infrastructure.
