ForgeGuardian

Supply chain security scanner for every package you depend on.
8 engines. 9 ecosystems. 223+ detection signatures. Works offline.

---

Install

One command. No account needed.

curl -sSfL https://raw.githubusercontent.com/Mah3Sec/ForgeGuardian/main/install.sh | bash

Windows (PowerShell)

irm https://raw.githubusercontent.com/Mah3Sec/ForgeGuardian/main/install.ps1 | iex

Docker

docker run -d --name forgeguardian -p 3000:3000 ghcr.io/mah3sec/forgeguardian

Open http://localhost:3000

Build from source

git clone https://github.com/Mah3Sec/ForgeGuardian.git
cd ForgeGuardian && make build

---

Scan

fgctl scan .

That's it. Scans your project, shows findings with severity and fix versions.

---

Dashboard

Start the web dashboard:

fgctl serve

Open http://localhost:8080 — SOC-style overview with security posture grading, severity trends, real-time alerts, and risk heatmaps.

Live demo: forgeguardian.mahendrapurbia.com

30+ pages across 7 categories

Analyze — Multi-engine vulnerability scanner (registry + file upload + remote SSH), scan session history with JSON/CSV/HTML export, recursive directory scanning, system audit across all package managers

AI-Powered — AI security analysis, AI advisory with remediation guidance, autonomous patch agent

Monitor — Live monitoring with auto-quarantine, structured log viewer, dependency topology with attack surface mapping, dependency drift detection, alert timeline

Inventory — Multi-workspace project management, dependency inventory with risk grades, SBOM generation (CycloneDX + SPDX)

Policy — Policy-as-code rules, allowlist/blocklist management, Sigstore keyless signing + verification, provenance tracking, signature authoring with guided wizard

Integrations — Webhook alerts (Slack, Discord, HTTP), CI/CD pipeline config, report exports

Tools — Built-in web terminal for CLI commands, developer docs, API reference, settings

---

8 Scan Engines

Every scan runs all available engines concurrently:

Engine	Status	What it catches
OSV	always runs	Known CVEs via osv.dev
Behavioral	always runs	Malicious install scripts, typosquatting
Malware	always runs	Byte/regex pattern matching
AI Model	always runs	Unsafe HuggingFace weights
MCP	always runs	Prompt injection in tool descriptions
Grype	optional	Deep CVE scan of artifacts
Trivy	optional	Container + OS scanning
Semgrep	optional	SAST static analysis

Missing an optional engine? Run fgctl doctor --fix to auto-install them.

---

9 Ecosystems

npm, PyPI, Go, Maven, RubyGems, Cargo/crates.io, NuGet, HuggingFace, GitHub Actions

---

223+ Community Signatures

ForgeGuardian ships with 223+ detection signatures covering real supply chain attacks (2016-2026):

Type	What it catches	Count
blocklisted_package	Confirmed malicious packages (event-stream, XZ utils, Shai-Hulud worm...)	80
behavioral_rule	Install-time env harvest, SSH key theft, dep confusion	42
malware_pattern	Obfuscated loaders, crypto miners, RATs, credential stealers	44
typosquatting_target	Popular packages + known typosquat variants	27
mcp_injection_pattern	Tool shadowing, data exfil via MCP	13
pickle_rule	Unsafe AI model weights, missing model cards	12

Update signatures:

fgctl update

Create your own:

fgctl intel new          # guided wizard
fgctl intel validate .   # validate schema
fgctl intel test .       # test against real package

---

CI/CD

Drop into any GitHub Actions workflow:

- name: Install ForgeGuardian
  run: |
    curl -sSfL https://raw.githubusercontent.com/Mah3Sec/ForgeGuardian/main/install.sh | bash
    echo "$HOME/.local/bin" >> $GITHUB_PATH

- name: Scan
  run: fgctl scan . --ci --fail-on=high --format=sarif > results.sarif

- name: Upload to GitHub Security
  uses: github/codeql-action/upload-sarif@v3
  with:
    sarif_file: results.sarif

---

Key Features

• Offline-first — all scans run locally, no data leaves your machine
• AI triage — optional AI advisory and patch agent (needs ANTHROPIC_API_KEY)
• Multi-workspace — organize projects into workspaces with independent scan histories
• SBOM — CycloneDX 1.5 + SPDX 2.3 generation
• Sigstore signing — keyless artifact signing + verification
• Policy-as-code — YAML policy rules, deny lists, threshold enforcement
• Attack surface mapping — dependency topology graph with risk visualization
• Webhooks — Slack, Discord, generic HTTP alerts
• Risk scoring — A-F letter grades per package
• Scan sessions — full history with JSON/CSV/HTML export
• Web terminal — run CLI commands from the dashboard
• Self-hostable — Docker one-liner, airgap-compatible
• SLSA Level 3 — provenance for every release

---

Free vs Pro

The engine, CLI, and community tools are Apache 2.0, free forever. Pro adds team features.

	Community (Free)	Pro
CLI scan + all 8 engines	Yes	Yes
SBOM, signing, provenance	Yes	Yes
Community signatures	Yes	Yes
Dashboard (self-hosted)	Yes	Yes
Alerts, policies, webhooks	Yes	Yes
AI advisory + patch agent	Yes	Yes
Team management + RBAC	—	Yes
Cloud-hosted option	—	Yes
SLA + priority support	—	Yes

---

Privacy

• Zero telemetry — phones home for nothing
• AI features are opt-in (explicit ANTHROPIC_API_KEY)
• Self-hostable and airgap-compatible
• SBOMs and provenance published for every release

---

Contributing

Fastest path: write a detection signature — no Go knowledge needed:

fgctl intel new    # guided wizard, ~10 minutes

For code contributions: fork, branch, PR. See CONTRIBUTING.md.

Security issues: SECURITY.md.

---

License

Apache License 2.0 — LICENSE