agent-bom-scan — AI Supply Chain Vulnerability Scanner
Checks packages for CVEs, scans container images natively, verifies package
provenance via Sigstore, scans filesystems, and generates SBOMs.
Install
pipx install agent-bom
agent-bom agents # discover agents and scan dependencies
agent-bom check langchain==0.1.0 # check a specific package with version
agent-bom image nginx:1.25 # scan container image (native)
agent-bom fs . # scan filesystem packages
agent-bom sbom . # generate SBOM
agent-bom verify agent-bom # verify Sigstore provenance
agent-bom where # show all discovery paths
As an MCP Server
{
"mcpServers": {
"agent-bom": {
"command": "uvx",
"args": ["agent-bom", "mcp"]
}
}
}
When to Use
- "check package" / "is this package safe"
- "scan image" / "scan container"
- "verify" / "check provenance"
- "is this safe" / "CVE lookup"
- "scan dependencies"
- "blast radius"
- "generate SBOM"
Tools (8)
| Tool |
Description |
check |
Check a package for CVEs (OSV, NVD, EPSS, KEV) |
scan |
Full discovery + vulnerability scan pipeline |
blast_radius |
Map CVE impact chain across agents, servers, credentials |
remediate |
Prioritized remediation plan for vulnerabilities |
verify |
Package integrity + SLSA provenance check |
diff |
Compare two scan reports (new/resolved/persistent) |
where |
Show MCP client config discovery paths |
inventory |
List discovered agents, servers, packages |
Examples
# Check a package before installing
check(package="langchain", version="0.1.0", ecosystem="pypi")
# Map blast radius of a CVE
blast_radius(cve_id="CVE-2024-21538")
# Full scan
scan()
# Verify package provenance
verify(package="agent-bom")
Guardrails
- Show CVEs even when NVD analysis is pending or severity is
unknown — a CVE ID is still a real finding.
- Treat
UNKNOWN severity as unresolved, not benign — it means data is not yet available.
- Do not modify any files, install packages, or change system configuration.
- Only public package names and CVE IDs leave the machine for vulnerability database lookups.
- Ask before scanning paths outside the user's home directory.
Privacy & Data Handling
# Step 1: Install
pip install agent-bom
# Step 2: Review redaction logic BEFORE scanning
# sanitize_env_vars() replaces ALL env var values with ***REDACTED***
# BEFORE any config data is processed or stored:
# https://github.com/msaad00/agent-bom/blob/main/src/agent_bom/security.py#L159
# Step 3: Verify package provenance (Sigstore)
agent-bom verify agent-bom
# Step 4: Only then run scans
agent-bom agents
Verification
- Source: github.com/msaad00/agent-bom (Apache-2.0)
- Sigstore signed:
agent-bom verify agent-bom@0.75.10
- 6,533+ tests with CodeQL + OpenSSF Scorecard
- No telemetry: Zero tracking, zero analytics
1---2name: agent-bom-scan3description: >-4 Open security platform for agentic infrastructure — checks packages for CVEs5 (OSV, NVD, EPSS, KEV), scans container images, verifies provenance, scans6 filesystems, and generates SBOMs. Use when: "check package", "scan image",7 "verify", "is this safe", "scan dependencies", "CVE lookup", "blast radius".8version: 0.75.109license: Apache-2.010compatibility: >-11 Requires Python 3.11+. Install via pipx or pip. Native container image12 scanning — no external scanner required. No API keys required for basic13 operation.14metadata:15 author: msaad0016 homepage: https://github.com/msaad00/agent-bom17 source: https://skillmd.com/skills/leoyeai-openclaw-master-skills/agent-bom-scan18 upstream: https://github.com/msaad00/agent-bom19 pypi: https://pypi.org/project/agent-bom/20 scorecard: https://securityscorecards.dev/viewer/?uri=github.com/msaad00/agent-bom21 tests: 653322 install:23 pipx: agent-bom24 pip: agent-bom25 docker: ghcr.io/msaad00/agent-bom:0.75.1026 openclaw:27 requires:28 bins: []29 env: []30 credentials: none31 credential_policy: "Zero credentials required. Optional env vars below increase rate limits. They are never auto-discovered, inferred, or transmitted."32 optional_env: []33 optional_bins:34 - semgrep35 - kubectl36 emoji: "\U0001F6E1"37 homepage: https://github.com/msaad00/agent-bom38 source: https://github.com/msaad00/agent-bom39 license: Apache-2.040 os:41 - darwin42 - linux43 - windows44 credential_handling: "Env var values are NEVER extracted from config files. sanitize_env_vars() replaces all env values with ***REDACTED*** BEFORE any config data is processed or stored. Only structural data (server names, commands, URLs) passes through. Source: https://github.com/msaad00/agent-bom/blob/main/src/agent_bom/security.py#L159"45 data_flow: "All scanning is local-first. Only public package names and CVE IDs are sent to vulnerability databases (OSV, NVD, EPSS, GitHub Advisories). No credentials, config file contents, or scan results leave the machine."46 file_reads:47 # Claude Desktop48 - "~/Library/Application Support/Claude/claude_desktop_config.json"49 - "~/.config/Claude/claude_desktop_config.json"50 # Claude Code51 - "~/.claude/settings.json"52 - "~/.claude.json"53 # Cursor54 - "~/.cursor/mcp.json"55 - "~/Library/Application Support/Cursor/User/globalStorage/cursor.mcp/mcp.json"56 # Windsurf57 - "~/.windsurf/mcp.json"58 # Cline59 - "~/Library/Application Support/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json"60 # VS Code Copilot61 - "~/Library/Application Support/Code/User/mcp.json"62 # Codex CLI63 - "~/.codex/config.toml"64 # Gemini CLI65 - "~/.gemini/settings.json"66 # Goose67 - "~/.config/goose/config.yaml"68 # Continue69 - "~/.continue/config.json"70 # Zed71 - "~/.config/zed/settings.json"72 # Roo Code73 - "~/Library/Application Support/Code/User/globalStorage/rooveterinaryinc.roo-cline/settings/cline_mcp_settings.json"74 # Amazon Q75 - "~/Library/Application Support/Code/User/globalStorage/amazonwebservices.amazon-q-vscode/mcp.json"76 # JetBrains AI77 - "~/Library/Application Support/JetBrains/*/mcp.json"78 - "~/.config/github-copilot/intellij/mcp.json"79 # Junie80 - "~/.junie/mcp/mcp.json"81 # GitHub Copilot CLI82 - "~/.copilot/mcp-config.json"83 # Tabnine84 - "~/.tabnine/mcp_servers.json"85 # Cortex Code (Snowflake)86 - "~/.snowflake/cortex/mcp.json"87 - "~/.snowflake/cortex/settings.json"88 - "~/.snowflake/cortex/permissions.json"89 - "~/.snowflake/cortex/hooks.json"90 # Snowflake CLI91 - "~/.snowflake/connections.toml"92 - "~/.snowflake/config.toml"93 # Project-level configs94 - ".mcp.json"95 - ".vscode/mcp.json"96 - ".cursor/mcp.json"97 # User-provided files98 - "user-provided SBOM files (CycloneDX/SPDX JSON)"99 file_writes: []100 network_endpoints:101 - url: "https://api.osv.dev/v1"102 purpose: "OSV vulnerability database — batch CVE lookup for packages"103 auth: false104 - url: "https://services.nvd.nist.gov/rest/json/cves/2.0"105 purpose: "NVD CVSS v4 enrichment — optional API key increases rate limit"106 auth: false107 - url: "https://api.first.org/data/v1/epss"108 purpose: "EPSS exploit probability scores"109 auth: false110 - url: "https://api.github.com/advisories"111 purpose: "GitHub Security Advisories — supplemental CVE lookup"112 auth: false113 telemetry: false114 persistence: false115 privilege_escalation: false116 always: false117 autonomous_invocation: restricted118---119120# agent-bom-scan — AI Supply Chain Vulnerability Scanner121122Checks packages for CVEs, scans container images natively, verifies package123provenance via Sigstore, scans filesystems, and generates SBOMs.124125## Install126127```bash128pipx install agent-bom129agent-bom agents # discover agents and scan dependencies130agent-bom check langchain==0.1.0 # check a specific package with version131agent-bom image nginx:1.25 # scan container image (native)132agent-bom fs . # scan filesystem packages133agent-bom sbom . # generate SBOM134agent-bom verify agent-bom # verify Sigstore provenance135agent-bom where # show all discovery paths136```137138### As an MCP Server139140```json141{142 "mcpServers": {143 "agent-bom": {144 "command": "uvx",145 "args": ["agent-bom", "mcp"]146 }147 }148}149```150151## When to Use152153- "check package" / "is this package safe"154- "scan image" / "scan container"155- "verify" / "check provenance"156- "is this safe" / "CVE lookup"157- "scan dependencies"158- "blast radius"159- "generate SBOM"160161## Tools (8)162163| Tool | Description |164|------|-------------|165| `check` | Check a package for CVEs (OSV, NVD, EPSS, KEV) |166| `scan` | Full discovery + vulnerability scan pipeline |167| `blast_radius` | Map CVE impact chain across agents, servers, credentials |168| `remediate` | Prioritized remediation plan for vulnerabilities |169| `verify` | Package integrity + SLSA provenance check |170| `diff` | Compare two scan reports (new/resolved/persistent) |171| `where` | Show MCP client config discovery paths |172| `inventory` | List discovered agents, servers, packages |173174## Examples175176```177# Check a package before installing178check(package="langchain", version="0.1.0", ecosystem="pypi")179180# Map blast radius of a CVE181blast_radius(cve_id="CVE-2024-21538")182183# Full scan184scan()185186# Verify package provenance187verify(package="agent-bom")188```189190## Guardrails191192- Show CVEs even when NVD analysis is pending or severity is `unknown` — a CVE ID is still a real finding.193- Treat `UNKNOWN` severity as unresolved, not benign — it means data is not yet available.194- Do not modify any files, install packages, or change system configuration.195- Only public package names and CVE IDs leave the machine for vulnerability database lookups.196- Ask before scanning paths outside the user's home directory.197198## Privacy & Data Handling199200```bash201# Step 1: Install202pip install agent-bom203204# Step 2: Review redaction logic BEFORE scanning205# sanitize_env_vars() replaces ALL env var values with ***REDACTED***206# BEFORE any config data is processed or stored:207# https://github.com/msaad00/agent-bom/blob/main/src/agent_bom/security.py#L159208209# Step 3: Verify package provenance (Sigstore)210agent-bom verify agent-bom211212# Step 4: Only then run scans213agent-bom agents214```215216## Verification217218- **Source**: [github.com/msaad00/agent-bom](https://github.com/msaad00/agent-bom) (Apache-2.0)219- **Sigstore signed**: `agent-bom verify agent-bom@0.75.10`220- **6,533+ tests** with CodeQL + OpenSSF Scorecard221- **No telemetry**: Zero tracking, zero analytics