convez.ai
MCP ServerSegurançaAvançado
msaad00

Agent-BOM: Scanner de Segurança para Infraestrutura de IA

msaad00/agent-bom

Inventário de segurança para agentes e MCP com análise de raio de explosão

Ver no GitHub
BomPluginScore · de 0 a 100

Scanner de segurança open-source e plano de controle auto-hospedado para agentes de IA e servidores MCP. Constrói um inventário completo (BOM) da infraestrutura de IA, mapeia vulnerabilidades através de grafo de dependências, expõe caminhos de ataque e fornece evidências de conformidade via CLI, API REST, ferramentas MCP e dashboard.

Instalação

$npx convez-ai msaad00/agent-bom

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
38
Forks
12
Acompanhando
0
Saúde
100%
Último commit
há 1 dia
Versão
v0.115.4
Issues abertas
21
Contribuidores
5

Casos de uso

  • Auditoria de segurança em infraestrutura de IA e agentes
  • Mapeamento de vulnerabilidades através de dependências de MCP servers
  • Conformidade e geração de evidências de segurança
  • Controle de acesso em tempo de execução para agentes em produção
  • Integração em pipelines CI/CD (GitHub Actions)

Veredito

Ferramenta robusta e bem-mantida para segurança em supply chain de IA. Preenche gap crítico ao conectar vulnerabilidades de pacotes a agentes específicos através de grafo de exposição. Recomendado para equipes com infraestrutura complexa de MCP/agentes que precisam de visibilidade real de risco.

README

A tradução para o português deste README ainda está na fila.

agent-bom scans local and fleet AI infrastructure, builds an AI BOM across agents, MCP servers, tools, packages, credential environment names, cloud, runtime, and skills, then turns that inventory into findings, compliance evidence, and graph-backed exposure paths.

The same evidence is available through CLI/CI, REST API, MCP tools, and a self-hosted dashboard. Runtime proxy/gateway controls are optional and scoped to environments where enforcement is worth the operational cost.

package
  -> vulnerability finding
  -> MCP server
  -> tools + credential refs
  -> agent

Blast radius is the core idea. A vulnerable package is not just a CVE row; it is linked to the MCP server that loads it, the tools exposed by that server, the credential environment names in reach, and the agents that can call it.

First Run

pip install agent-bom
agent-bom quickstart --dry-run --offline   # print the onboarding plan
agent-bom quickstart --run --offline        # write sample, scan, seed gateway policy, populate the cockpit
agent-bom agents --demo --offline

The demo uses real OSV/GHSA advisories against intentionally vulnerable sample packages and produces graph-ready inventory without touching your source tree. For a real local scan:

agent-bom agents -p . -f html -o agent-bom-report.html

Want an inspectable sample stack first?

agent-bom samples first-run
agent-bom agents --inventory agent-bom-first-run/inventory.json -p agent-bom-first-run --enrich

See docs/FIRST_RUN.md for the guided path from CLI output to the dashboard.

To reproduce the dashboard screenshots from a clean local control-plane store:

make build-ui
uv run agent-bom serve --persist /tmp/agent-bom-demo.db --allow-insecure-no-auth
uv run agent-bom agents --demo --offline --no-auto-update-db -f json -o /tmp/agent-bom-demo.json
curl -sS -H 'content-type: application/json' --data-binary @/tmp/agent-bom-demo.json \
  http://127.0.0.1:8422/v1/results/push

Product Proof

The dashboard screenshots below are captured from the packaged UI with bundled demo scan data and seeded control-plane records, not static mockups. The data is synthetic where needed, but the routes are the real scan, graph, fleet, identity, audit, and gateway surfaces. The README keeps the first screen focused; expand the gallery when you want to inspect the control-plane surfaces.

The graph proof set is intentionally split across modes: fix-first exposure paths, root-centered lineage, lateral context, and package risk distribution. That keeps each view readable instead of forcing every relationship into one sprawling canvas.

Fleet and identity views use the same control-plane APIs that operators use for customer-owned deployments. The sample below seeds environment, owner, lifecycle state, and agent identity events so the screenshots show how local scan evidence connects to reviewable governance records.

Screenshot capture rules and the full manifest live in docs/CAPTURE.md and docs/images/product-screenshots.json.

Start Here

GoalCommandArtifact
Local agent and MCP inventoryagent-bom agentsfindings, AI BOM, graph-ready JSON
Guided local onboardingagent-bom quickstart --dry-run --offlinescan, sample-data, and local API/UI next steps
One-command onboardingagent-bom quickstart --run --offlinewrites sample, runs a graph-persisting scan, seeds a baseline gateway policy
Repo and lockfile scanagent-bom agents -p .package findings, SARIF/SBOM/HTML when requested
Pre-install guardagent-bom check flask@2.0.0 --ecosystem pypideterministic allow/warn/block result
Container image scanagent-bom image nginx:latestimage findings and remediation
IaC scanagent-bom iac Dockerfile k8s/ infra/main.tfIaC findings and policy context
Cloud posture checkagent-bom cis-benchmark --provider awsruntime CIS posture evidence
CI gateuses: msaad00/agent-bom@v0.88.6SARIF, PR summary, optional code-scanning upload
MCP toolspip install 'agent-bom[mcp-server]' && agent-bom mcp serverstrict-args security tools for MCP clients
Local API/UIpip install 'agent-bom[ui]' && agent-bom serveAPI plus bundled dashboard
First-run extraspip install 'agent-bom[all]'supported onboarding extras; MLflow remains separately installed
Self-hosted pilotdocker compose -f docker-compose.pilot.yml up -dAPI and dashboard in your environment

The base wheel is the scanner and CLI path. Optional runtime surfaces fail fast with install hints when their extras are missing.

MCP registry publishing is tracked through the committed Smithery manifest and other registry metadata; install and liveness checks stay in the linked integration docs instead of this front door.

Shipped Surfaces

SurfacePrimary userCurrent boundary
CLI / CIdevelopers and release gateslocal scans, SARIF/SBOM/HTML/JSON, deterministic exit codes
REST APIcontrol-plane integrationsscans, bulk findings, dataset versions, evaluation runs, graph evidence, audit, runtime summaries
MCP toolsagents and assistantsstrict arguments, read-mostly security queries, exposure paths, deploy decisions, audited Shield actions
Dashboardsecurity teams and operatorsinventory, findings, graph cockpit, compliance, evidence, runtime posture
Runtime proxy/gatewayruntime operatorsscoped MCP traffic inspection, policy decisions, redacted audit evidence
Python clientservices, notebooks, and automationtyped helper for stable REST endpoints in the packaged wheel
TypeScript clientservices and agent runtimestyped helper for stable REST endpoints

MCP server mode advertises 63 MCP tools, 6 resources, and 6 workflow prompts. Most tools are read-only. The three Shield write actions fail closed unless the caller supplies operator_role=admin, operator_scopes=shield:write, and an audit reason.

CLI scan commands run local scan pipelines today. They share lower scanner and discovery libraries with the API, but they are not API wrappers yet.

Runtime enforcement is explicit. Proxy mode either wraps a target MCP server for audit and policy decisions, or runs that server through Docker/Podman isolation when a sandbox image is supplied:

agent-bom proxy --no-isolate --policy policy.json --detect-credentials --block-undeclared -- npx @mcp/server-github
agent-bom proxy --sandbox-image ghcr.io/acme/mcp-runtime@sha256:<digest> \
  --sandbox-image-pin-policy enforce --block-undeclared -- npx @mcp/server-postgres

Deploy In Your Boundary

agent-bom is designed for customer-controlled deployment: local CLI, Docker, GitHub Action, Helm, EKS, Postgres, and optional runtime proxy/gateway.

curl -fsSL https://raw.githubusercontent.com/msaad00/agent-bom/main/deploy/docker-compose.pilot.yml -o docker-compose.pilot.yml
docker compose -f docker-compose.pilot.yml up -d
# Dashboard -> http://localhost:3000

Production self-hosting starts with the deployment chooser:

There is no managed cloud offering in this repository today. Product lane boundaries are documented in docs/PRODUCT_BOUNDARIES.md.

Trust Model

  • Read-only discovery by default for cloud and local inventory.
  • No mandatory telemetry.
  • Credential values are redacted; credential environment names are preserved as evidence so exposure paths stay explainable.
  • Findings can export as JSON, SARIF, CycloneDX, SPDX, Markdown, HTML, and compliance evidence bundles.
  • API and runtime paths are designed for tenant scope, auth boundaries, and audit evidence.
  • OpenAPI artifacts are committed for SDK and client contract checks.

Security and release references:

Product Views

The docs site carries the deployment-oriented walkthroughs behind those screenshots:

Contributing

Contributions are welcome. Start with:

License: Apache-2.0.

Ver o original no GitHub

Bom para

Relacionados