agent-bom — AI Agent Infrastructure Security Scanner
Discovers MCP clients and servers across 22 AI tools, scans for CVEs, maps
blast radius, runs cloud CIS benchmarks, checks OWASP/NIST/MITRE compliance,
generates SBOMs, and assesses AI infrastructure against AISVS v1.0 and MAESTRO
framework layers.
Install
pipx install agent-bom
agent-bom agents # auto-discover + scan
agent-bom check langchain==0.1.0 # check a specific package with version
agent-bom fs . # scan filesystem packages
agent-bom image nginx:1.25 # scan container image (native)
agent-bom cloud aws # AWS CIS benchmark
agent-bom iac infra/ # scan Terraform/CloudFormation
agent-bom where # show all discovery paths
As an MCP Server
{
"mcpServers": {
"agent-bom": {
"command": "uvx",
"args": ["agent-bom", "mcp"]
}
}
}
Sub-Skills (8)
| Sub-Skill |
Purpose |
Triggers |
| discover |
Find agents, MCP servers, configurations |
"find agents", "what's configured", "mcp inventory" |
| scan |
CVE scanning, image scanning, SBOM, provenance |
"check package", "scan image", "verify", "blast radius" |
| scan-infra |
IaC, cloud config, secrets scanning |
"check terraform", "scan kubernetes", "find secrets" |
| enforce |
Runtime policy enforcement, MCP proxy |
"block risky calls", "apply policy", "proxy" |
| compliance |
11-framework compliance, SBOM generation |
"compliance report", "NIST", "SOC 2", "OWASP" |
| monitor |
Fleet monitoring, trust scores, lifecycle |
"fleet", "watch agents", "trust scores" |
| analyze |
Blast radius, attack paths, context graph |
"blast radius", "threat intel", "attack path" |
| troubleshoot |
Diagnostics, doctor, config validation |
"doctor", "debug", "why failing", "validate config" |
Tools (32)
Vulnerability Scanning
| Tool |
Description |
scan |
Full discovery + vulnerability scan pipeline |
check |
Check a package for CVEs (OSV, NVD, EPSS, KEV) |
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 |
Compliance & Policy
| Tool |
Description |
compliance |
OWASP LLM/Agentic Top 10, EU AI Act, MITRE ATLAS, NIST AI RMF |
policy_check |
Evaluate results against custom security policy (17 conditions) |
cis_benchmark |
CIS benchmark checks (AWS, Azure v3.0, GCP v3.0, Snowflake) |
generate_sbom |
Generate SBOM (CycloneDX or SPDX format) |
aisvs_benchmark |
OWASP AISVS v1.0 compliance — 9 AI security checks |
Registry & Trust
| Tool |
Description |
registry_lookup |
Look up MCP server in 427+ server security metadata registry |
marketplace_check |
Pre-install trust check with registry cross-reference |
fleet_scan |
Batch registry lookup + risk scoring for MCP server inventories |
skill_trust |
Assess skill file trust level (5-category analysis) |
code_scan |
SAST scanning via Semgrep with CWE-based compliance mapping |
Runtime & Analytics
| Tool |
Description |
context_graph |
Agent context graph with lateral movement analysis |
analytics_query |
Query vulnerability trends, posture history, and runtime events |
runtime_correlate |
Cross-reference proxy audit JSONL with CVE findings, risk amplification |
vector_db_scan |
Probe Qdrant/Weaviate/Chroma/Milvus for auth and exposure |
gpu_infra_scan |
GPU container and K8s node inventory + unauthenticated DCGM probe (MAESTRO KC6) |
Specialized Scans
| Tool |
Description |
dataset_card_scan |
Scan dataset cards for bias, licensing, and provenance issues |
training_pipeline_scan |
Scan training pipeline configs for security risks |
browser_extension_scan |
Scan browser extensions for risky permissions and AI domain access |
model_provenance_scan |
Verify model provenance and supply chain integrity |
prompt_scan |
Scan prompt templates for injection and data leakage risks |
model_file_scan |
Scan model files for unsafe serialization (pickle, etc.) |
license_compliance_scan |
Full SPDX license catalog scan with copyleft and network-copyleft detection |
ingest_external_scan |
Import external scan results (CycloneDX/SPDX/JSON) and merge into agent-bom findings |
Resources
| Resource |
Description |
registry://servers |
Browse 427+ MCP server security metadata registry |
Example Workflows
# Check a package before installing
check(package="@modelcontextprotocol/server-filesystem", ecosystem="npm")
# Map blast radius of a CVE
blast_radius(cve_id="CVE-2024-21538")
# Full agent discovery + scan
agents()
# Run CIS benchmark
cis_benchmark(provider="aws")
# Run AISVS v1.0 compliance
aisvs_benchmark()
# Scan vector databases for auth misconfigurations
vector_db_scan()
# Discover GPU containers, K8s GPU nodes, and unauthenticated DCGM endpoints
gpu_infra_scan()
# Assess trust of a skill file
skill_trust(skill_content="<paste SKILL.md content>")
Guardrails
Always do:
- Show CVEs even when NVD analysis is pending or severity is
unknown — a CVE ID with no details is still a real finding. Report what is known; mark severity as unknown explicitly.
- Confirm with the user before scanning cloud environments (
cis_benchmark) — these make live API calls to AWS/Azure/GCP using the user's credentials.
- Treat
UNKNOWN severity as unresolved, not benign — it means data is not yet available, not that the issue is minor.
Never do:
- Do not modify any files, install packages, or change system configuration. This skill is read-only.
- Do not transmit env var values, credentials, or file contents to any external service. Only package names and CVE IDs leave the machine.
- Do not invoke
agents() autonomously on sensitive environments without user confirmation. The autonomous_invocation policy is restricted.
Stop and ask the user when:
- The user requests a cloud CIS benchmark and no cloud credentials are configured.
- A scan finds
CRITICAL CVEs — present findings and ask whether to generate a remediation plan.
- The user asks to scan a path outside their home directory.
Supported Frameworks (14)
- OWASP LLM Top 10 (2025) — prompt injection, supply chain, data leakage
- OWASP MCP Top 10 — MCP-specific security risks
- OWASP Agentic Top 10 — tool poisoning, rug pulls, credential theft
- OWASP AISVS v1.0 — AI Security Verification Standard (9 checks)
- MITRE ATLAS — adversarial ML threat framework
- NIST AI RMF — govern, map, measure, manage lifecycle
- NIST CSF 2.0 — identify, protect, detect, respond, recover
- NIST 800-53 Rev 5 — federal security controls (CM-8, RA-5, SI-2, SR-3)
- FedRAMP Moderate — derived from NIST 800-53 controls
- EU AI Act — risk classification, transparency, SBOM requirements
- ISO 27001:2022 — information security controls (Annex A)
- SOC 2 — Trust Services Criteria
- CIS Controls v8 — implementation groups IG1/IG2/IG3
- CMMC 2.0 — cybersecurity maturity model (Level 1-3)
Privacy & Data Handling
This skill installs agent-bom from PyPI. Verify the redaction behavior
before running with any config files:
# 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: Review config parsing — only structural data extracted:
# https://github.com/msaad00/agent-bom/blob/main/src/agent_bom/discovery/__init__.py
# Step 4: Verify package provenance (Sigstore)
agent-bom verify agent-bom
# Step 5: Only then run scans
agent-bom agents
What is extracted: Server names, commands, args, and URLs from MCP client
config files across 22 AI tools. What is NOT extracted: Env var values are
replaced with ***REDACTED*** by sanitize_env_vars() before any processing.
Only public package names and CVE IDs are sent to vulnerability databases.
Cloud CIS checks use locally configured credentials and call only the cloud
provider's own APIs.
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-bom3description: >-4 Open security platform for agentic infrastructure — broad scanning plus MCP5 discovery, CVEs, blast radius, SBOMs, CIS benchmarks (AWS, Azure, GCP,6 Snowflake), OWASP/NIST/MITRE compliance, AISVS v1.0, MAESTRO layer tagging,7 and vector database security checks. Use when the user mentions vulnerability8 scanning, MCP server trust, compliance, SBOM generation, CIS benchmarks,9 blast radius, or AI supply chain risk.10version: 0.75.1011license: Apache-2.012compatibility: >-13 Requires Python 3.11+. Install via pipx or pip. No credentials required for14 basic scanning. Native container image scanning — no external scanner required.15 CIS benchmark checks optionally use cloud SDK credentials16 (AWS/Azure/GCP/Snowflake).17metadata:18 author: msaad0019 homepage: https://github.com/msaad00/agent-bom20 source: https://skillmd.com/skills/leoyeai-openclaw-master-skills/agent-bom21 upstream: https://github.com/msaad00/agent-bom22 pypi: https://pypi.org/project/agent-bom/23 scorecard: https://securityscorecards.dev/viewer/?uri=github.com/msaad00/agent-bom24 tests: 653325 install:26 pipx: agent-bom27 pip: agent-bom28 docker: ghcr.io/msaad00/agent-bom:0.75.1029 openclaw:30 requires:31 bins: []32 env: []33 credentials: none34 credential_policy: >-35 Zero credentials required for CVE scanning, blast radius, compliance36 evaluation, SBOM generation, and MCP registry lookups. Optional env vars37 below increase rate limits or enable cloud CIS checks. Env var values in38 discovered config files are replaced with ***REDACTED*** by39 sanitize_env_vars() in the installed code — verify at40 https://github.com/msaad00/agent-bom/blob/main/src/agent_bom/security.py#L15941 optional_env:42 - name: SNYK_TOKEN43 purpose: "Snyk vulnerability enrichment for code_scan (optional)"44 required: false45 - name: AWS_PROFILE46 purpose: "AWS CIS benchmark checks — uses boto3 with local AWS profile"47 required: false48 - name: AZURE_TENANT_ID49 purpose: "Azure CIS benchmark checks (azure-mgmt-* SDK)"50 required: false51 - name: AZURE_CLIENT_ID52 purpose: "Azure CIS benchmark checks — service principal client ID"53 required: false54 - name: AZURE_CLIENT_SECRET55 purpose: "Azure CIS benchmark checks — service principal secret"56 required: false57 - name: GOOGLE_APPLICATION_CREDENTIALS58 purpose: "GCP CIS benchmark checks (google-cloud-* SDK)"59 required: false60 - name: SNOWFLAKE_ACCOUNT61 purpose: "Snowflake CIS benchmark checks"62 required: false63 - name: SNOWFLAKE_USER64 purpose: "Snowflake CIS benchmark checks"65 required: false66 - name: SNOWFLAKE_PRIVATE_KEY_PATH67 purpose: "Snowflake key-pair auth (CI/CD)"68 required: false69 - name: SNOWFLAKE_AUTHENTICATOR70 purpose: "Snowflake auth method (default: externalbrowser SSO)"71 required: false72 optional_bins:73 - semgrep74 - kubectl75 emoji: "\U0001F6E1"76 homepage: https://github.com/msaad00/agent-bom77 source: https://github.com/msaad00/agent-bom78 license: Apache-2.079 os:80 - darwin81 - linux82 - windows83 install_verification: >-84 VERIFY BEFORE running with any config files: (1) pip install agent-bom;85 (2) Review sanitize_env_vars() at security.py#L159 — confirms env value86 redaction; (3) Review discovery/__init__.py — confirms only structural87 config data extracted; (4) agent-bom verify agent-bom — Sigstore88 provenance check; (5) Only then run agent-bom agents89 credential_handling: >-90 Env var values are NEVER extracted from config files. sanitize_env_vars()91 replaces all env values with ***REDACTED*** BEFORE any config data is92 processed or stored. Only structural data (server names, commands, URLs)93 passes through. Source:94 https://github.com/msaad00/agent-bom/blob/main/src/agent_bom/security.py#L15995 data_flow: >-96 Scanning is local-first. What leaves the machine: (1) public package names97 and CVE IDs sent to vulnerability databases (OSV, NVD, EPSS, GitHub98 Advisories) for CVE lookup; (2) CIS benchmark checks make read-only API99 calls to cloud providers (AWS/Azure/GCP/Snowflake) using your locally100 configured credentials, only when explicitly invoked. What stays local:101 all config file contents, env var values, credentials, scan results,102 compliance tags, and SBOM data. Registry lookups (427+ MCP servers) are103 bundled in-package with zero network calls. Env var values in discovered104 config files are replaced with ***REDACTED*** by sanitize_env_vars() in105 the installed code.106 file_reads:107 # Claude Desktop108 - "~/Library/Application Support/Claude/claude_desktop_config.json"109 - "~/.config/Claude/claude_desktop_config.json"110 # Claude Code111 - "~/.claude/settings.json"112 - "~/.claude.json"113 # Cursor114 - "~/.cursor/mcp.json"115 - "~/Library/Application Support/Cursor/User/globalStorage/cursor.mcp/mcp.json"116 # Windsurf117 - "~/.windsurf/mcp.json"118 # Cline119 - "~/Library/Application Support/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json"120 # VS Code Copilot121 - "~/Library/Application Support/Code/User/mcp.json"122 # Codex CLI123 - "~/.codex/config.toml"124 # Gemini CLI125 - "~/.gemini/settings.json"126 # Goose127 - "~/.config/goose/config.yaml"128 # Continue129 - "~/.continue/config.json"130 # Zed131 - "~/.config/zed/settings.json"132 # Roo Code133 - "~/Library/Application Support/Code/User/globalStorage/rooveterinaryinc.roo-cline/settings/cline_mcp_settings.json"134 # Amazon Q135 - "~/Library/Application Support/Code/User/globalStorage/amazonwebservices.amazon-q-vscode/mcp.json"136 # JetBrains AI137 - "~/Library/Application Support/JetBrains/*/mcp.json"138 - "~/.config/github-copilot/intellij/mcp.json"139 # Junie140 - "~/.junie/mcp/mcp.json"141 # GitHub Copilot CLI142 - "~/.copilot/mcp-config.json"143 # Tabnine144 - "~/.tabnine/mcp_servers.json"145 # Cortex Code (Snowflake)146 - "~/.snowflake/cortex/mcp.json"147 - "~/.snowflake/cortex/settings.json"148 - "~/.snowflake/cortex/permissions.json"149 - "~/.snowflake/cortex/hooks.json"150 # Snowflake CLI151 - "~/.snowflake/connections.toml"152 - "~/.snowflake/config.toml"153 # Project-level configs154 - ".mcp.json"155 - ".vscode/mcp.json"156 - ".cursor/mcp.json"157 # User-provided files158 - "user-provided SBOM files (CycloneDX/SPDX JSON)"159 - "user-provided policy files (YAML/JSON policy-as-code)"160 - "user-provided audit log files (JSONL from agent-bom proxy)"161 - "user-provided SKILL.md files (for skill_trust analysis)"162 file_writes: []163 network_endpoints:164 - url: "https://api.osv.dev/v1"165 purpose: "OSV vulnerability database — batch CVE lookup for packages"166 auth: false167 - url: "https://services.nvd.nist.gov/rest/json/cves/2.0"168 purpose: "NVD secondary enrichment — adds CWE IDs, dates, references (no key required)"169 auth: false170 - url: "https://api.first.org/data/v1/epss"171 purpose: "EPSS exploit probability scores"172 auth: false173 - url: "https://api.github.com/advisories"174 purpose: "GitHub Security Advisories — supplemental CVE lookup"175 auth: false176 - url: "https://api.snyk.io"177 purpose: "Snyk vulnerability enrichment for code_scan (requires SNYK_TOKEN)"178 auth: true179 - url: "https://*.amazonaws.com"180 purpose: "AWS CIS benchmark checks — read-only API calls (optional, user-initiated)"181 auth: true182 optional: true183 - url: "https://management.azure.com"184 purpose: "Azure CIS benchmark checks — read-only API calls (optional, user-initiated)"185 auth: true186 optional: true187 - url: "https://*.googleapis.com"188 purpose: "GCP CIS benchmark checks — read-only API calls (optional, user-initiated)"189 auth: true190 optional: true191 - url: "https://*.snowflakecomputing.com"192 purpose: "Snowflake CIS benchmark checks — read-only API calls (optional, user-initiated)"193 auth: true194 optional: true195 telemetry: false196 persistence: false197 privilege_escalation: false198 always: false199 autonomous_invocation: restricted200---201202# agent-bom — AI Agent Infrastructure Security Scanner203204Discovers MCP clients and servers across 22 AI tools, scans for CVEs, maps205blast radius, runs cloud CIS benchmarks, checks OWASP/NIST/MITRE compliance,206generates SBOMs, and assesses AI infrastructure against AISVS v1.0 and MAESTRO207framework layers.208209## Install210211```bash212pipx install agent-bom213agent-bom agents # auto-discover + scan214agent-bom check langchain==0.1.0 # check a specific package with version215agent-bom fs . # scan filesystem packages216agent-bom image nginx:1.25 # scan container image (native)217agent-bom cloud aws # AWS CIS benchmark218agent-bom iac infra/ # scan Terraform/CloudFormation219agent-bom where # show all discovery paths220```221222### As an MCP Server223224```json225{226 "mcpServers": {227 "agent-bom": {228 "command": "uvx",229 "args": ["agent-bom", "mcp"]230 }231 }232}233```234235## Sub-Skills (8)236237| Sub-Skill | Purpose | Triggers |238|-----------|---------|---------|239| [discover](discover/SKILL.md) | Find agents, MCP servers, configurations | "find agents", "what's configured", "mcp inventory" |240| [scan](scan/SKILL.md) | CVE scanning, image scanning, SBOM, provenance | "check package", "scan image", "verify", "blast radius" |241| [scan-infra](scan-infra/SKILL.md) | IaC, cloud config, secrets scanning | "check terraform", "scan kubernetes", "find secrets" |242| [enforce](enforce/SKILL.md) | Runtime policy enforcement, MCP proxy | "block risky calls", "apply policy", "proxy" |243| [compliance](compliance/SKILL.md) | 11-framework compliance, SBOM generation | "compliance report", "NIST", "SOC 2", "OWASP" |244| [monitor](monitor/SKILL.md) | Fleet monitoring, trust scores, lifecycle | "fleet", "watch agents", "trust scores" |245| [analyze](analyze/SKILL.md) | Blast radius, attack paths, context graph | "blast radius", "threat intel", "attack path" |246| [troubleshoot](troubleshoot/SKILL.md) | Diagnostics, doctor, config validation | "doctor", "debug", "why failing", "validate config" |247248## Tools (32)249250### Vulnerability Scanning251| Tool | Description |252|------|-------------|253| `scan` | Full discovery + vulnerability scan pipeline |254| `check` | Check a package for CVEs (OSV, NVD, EPSS, KEV) |255| `blast_radius` | Map CVE impact chain across agents, servers, credentials |256| `remediate` | Prioritized remediation plan for vulnerabilities |257| `verify` | Package integrity + SLSA provenance check |258| `diff` | Compare two scan reports (new/resolved/persistent) |259| `where` | Show MCP client config discovery paths |260| `inventory` | List discovered agents, servers, packages |261262### Compliance & Policy263| Tool | Description |264|------|-------------|265| `compliance` | OWASP LLM/Agentic Top 10, EU AI Act, MITRE ATLAS, NIST AI RMF |266| `policy_check` | Evaluate results against custom security policy (17 conditions) |267| `cis_benchmark` | CIS benchmark checks (AWS, Azure v3.0, GCP v3.0, Snowflake) |268| `generate_sbom` | Generate SBOM (CycloneDX or SPDX format) |269| `aisvs_benchmark` | OWASP AISVS v1.0 compliance — 9 AI security checks |270271### Registry & Trust272| Tool | Description |273|------|-------------|274| `registry_lookup` | Look up MCP server in 427+ server security metadata registry |275| `marketplace_check` | Pre-install trust check with registry cross-reference |276| `fleet_scan` | Batch registry lookup + risk scoring for MCP server inventories |277| `skill_trust` | Assess skill file trust level (5-category analysis) |278| `code_scan` | SAST scanning via Semgrep with CWE-based compliance mapping |279280### Runtime & Analytics281| Tool | Description |282|------|-------------|283| `context_graph` | Agent context graph with lateral movement analysis |284| `analytics_query` | Query vulnerability trends, posture history, and runtime events |285| `runtime_correlate` | Cross-reference proxy audit JSONL with CVE findings, risk amplification |286| `vector_db_scan` | Probe Qdrant/Weaviate/Chroma/Milvus for auth and exposure |287| `gpu_infra_scan` | GPU container and K8s node inventory + unauthenticated DCGM probe (MAESTRO KC6) |288289### Specialized Scans290| Tool | Description |291|------|-------------|292| `dataset_card_scan` | Scan dataset cards for bias, licensing, and provenance issues |293| `training_pipeline_scan` | Scan training pipeline configs for security risks |294| `browser_extension_scan` | Scan browser extensions for risky permissions and AI domain access |295| `model_provenance_scan` | Verify model provenance and supply chain integrity |296| `prompt_scan` | Scan prompt templates for injection and data leakage risks |297| `model_file_scan` | Scan model files for unsafe serialization (pickle, etc.) |298| `license_compliance_scan` | Full SPDX license catalog scan with copyleft and network-copyleft detection |299| `ingest_external_scan` | Import external scan results (CycloneDX/SPDX/JSON) and merge into agent-bom findings |300301### Resources302| Resource | Description |303|----------|-------------|304| `registry://servers` | Browse 427+ MCP server security metadata registry |305306## Example Workflows307308```309# Check a package before installing310check(package="@modelcontextprotocol/server-filesystem", ecosystem="npm")311312# Map blast radius of a CVE313blast_radius(cve_id="CVE-2024-21538")314315# Full agent discovery + scan316agents()317318# Run CIS benchmark319cis_benchmark(provider="aws")320321# Run AISVS v1.0 compliance322aisvs_benchmark()323324# Scan vector databases for auth misconfigurations325vector_db_scan()326327# Discover GPU containers, K8s GPU nodes, and unauthenticated DCGM endpoints328gpu_infra_scan()329330# Assess trust of a skill file331skill_trust(skill_content="<paste SKILL.md content>")332```333334## Guardrails335336**Always do:**337- Show CVEs even when NVD analysis is pending or severity is `unknown` — a CVE ID with no details is still a real finding. Report what is known; mark severity as `unknown` explicitly.338- Confirm with the user before scanning cloud environments (`cis_benchmark`) — these make live API calls to AWS/Azure/GCP using the user's credentials.339- Treat `UNKNOWN` severity as unresolved, not benign — it means data is not yet available, not that the issue is minor.340341**Never do:**342- Do not modify any files, install packages, or change system configuration. This skill is read-only.343- Do not transmit env var values, credentials, or file contents to any external service. Only package names and CVE IDs leave the machine.344- Do not invoke `agents()` autonomously on sensitive environments without user confirmation. The `autonomous_invocation` policy is `restricted`.345346**Stop and ask the user when:**347- The user requests a cloud CIS benchmark and no cloud credentials are configured.348- A scan finds `CRITICAL` CVEs — present findings and ask whether to generate a remediation plan.349- The user asks to scan a path outside their home directory.350351## Supported Frameworks (14)352353- **OWASP LLM Top 10** (2025) — prompt injection, supply chain, data leakage354- **OWASP MCP Top 10** — MCP-specific security risks355- **OWASP Agentic Top 10** — tool poisoning, rug pulls, credential theft356- **OWASP AISVS v1.0** — AI Security Verification Standard (9 checks)357- **MITRE ATLAS** — adversarial ML threat framework358- **NIST AI RMF** — govern, map, measure, manage lifecycle359- **NIST CSF 2.0** — identify, protect, detect, respond, recover360- **NIST 800-53 Rev 5** — federal security controls (CM-8, RA-5, SI-2, SR-3)361- **FedRAMP Moderate** — derived from NIST 800-53 controls362- **EU AI Act** — risk classification, transparency, SBOM requirements363- **ISO 27001:2022** — information security controls (Annex A)364- **SOC 2** — Trust Services Criteria365- **CIS Controls v8** — implementation groups IG1/IG2/IG3366- **CMMC 2.0** — cybersecurity maturity model (Level 1-3)367368## Privacy & Data Handling369370This skill installs agent-bom from PyPI. **Verify the redaction behavior371before running with any config files:**372373```bash374# Step 1: Install375pip install agent-bom376377# Step 2: Review redaction logic BEFORE scanning378# sanitize_env_vars() replaces ALL env var values with ***REDACTED***379# BEFORE any config data is processed or stored:380# https://github.com/msaad00/agent-bom/blob/main/src/agent_bom/security.py#L159381382# Step 3: Review config parsing — only structural data extracted:383# https://github.com/msaad00/agent-bom/blob/main/src/agent_bom/discovery/__init__.py384385# Step 4: Verify package provenance (Sigstore)386agent-bom verify agent-bom387388# Step 5: Only then run scans389agent-bom agents390```391392**What is extracted**: Server names, commands, args, and URLs from MCP client393config files across 22 AI tools. **What is NOT extracted**: Env var values are394replaced with `***REDACTED***` by `sanitize_env_vars()` before any processing.395Only public package names and CVE IDs are sent to vulnerability databases.396Cloud CIS checks use locally configured credentials and call only the cloud397provider's own APIs.398399## Verification400401- **Source**: [github.com/msaad00/agent-bom](https://github.com/msaad00/agent-bom) (Apache-2.0)402- **Sigstore signed**: `agent-bom verify agent-bom@0.75.10`403- **6,533+ tests** with CodeQL + OpenSSF Scorecard404- **No telemetry**: Zero tracking, zero analytics