Local workflows

Every prompt on this server drives your machine rather than the server. They need a repository and the Vulnetix CLI.

That is not a subset any more. There used to be seventeen that ran server-side, and each of them restated a tool sitting next to it — so they were removed, and what remains is only the work a server genuinely cannot do.

Why the split exists

The server is a Cloudflare Worker. It has no filesystem, cannot run a package manager, and cannot invoke Semgrep or Trivy. That is not a gap waiting to be filled. It is the reason no code leaves your machine.

The alternative would be uploading your source to be scanned. Every one of these workflows would then require shipping the thing you are trying to protect to a third party. The boundary is the guarantee.

Install the CLI

curl -fsSL https://cli.vulnetix.com/install.sh | sh

Or by package manager:

brew install vulnetix/tap/vulnetix          # macOS / Linuxbrew
scoop bucket add vulnetix https://github.com/Vulnetix/scoop-bucket
scoop install vulnetix                       # Windows
nix profile install github:Vulnetix/cli      # Nix
go install github.com/Vulnetix/cli/cmd/vulnetix@latest

Then authenticate once, with the same credential the MCP server uses:

vulnetix auth login
vulnetix auth status

Full documentation: docs.cli.vulnetix.com.

What each one needs

PromptNeeds
vulnetix_sast_scanRepo. Optional: Semgrep + .semgrep config
vulnetix_secret_scanRepo, or staged changes
vulnetix_iac_scan.tf / .tofu / k8s manifests
vulnetix_container_scanDockerfile. Optional: Trivy, Grype, Syft, Docker
vulnetix_license_checkDependency tree
vulnetix_sbom_generateRepo. Optional: Syft, cosign
vulnetix_eol_checkManifests
vulnetix_fixWritable manifests, a package manager
vulnetix_dep_resolvePackage manager (npm ls, cargo tree, …)
vulnetix_verify_fixRepo, for the re-scan gate
vulnetix_vex_publish.vulnetix/memory.yaml. Optional: gh
vulnetix_dependency_choiceManifests, and a conversation with you
vulnetix_repo_impactRepo, for the dependency graph
vulnetix_typosquat_checkManifests
vulnetix_detection_rulesSomewhere to write the rules
vulnetix_exploit_testAn authorised target. Optional: Nuclei
vulnetix_secure_code_writeThe file you are editing
vulnetix_dashboard.vulnetix/memory.yaml only

Third-party binaries are optional augmentation, never requirements. The CLI has its own scanners; Semgrep, Trivy, Grype, Syft and cosign compose with them when present. vulnetix agent capabilities writes .vulnetix/capabilities.yaml recording what it found, and these workflows read it to decide what to offer. That detection was a prompt until it became a CLI command, which is also when it started working on Windows.

Which agents can run these

The prompt tells your agent to run shell commands, so the agent needs to be able to.

ClientCan run them
Claude CodeYes, Bash tool
Cursor, WindsurfYes, agent mode terminal
VS Code / CopilotYes, agent mode terminal
Cline, Roo CodeYes
Continue, GooseYes
ZedYes, agent panel terminal
Claude DesktopNo, no shell and no repository

In Claude Desktop these prompts still render, and are still readable as a procedure, but nothing will execute. Use a terminal-based agent.

Shared state

Several workflows read and write .vulnetix/memory.yaml in your repository, holding tracked findings, triage decisions and history. It is the CLI’s file, and the MCP server never sees it.

vulnetix_dashboard reads it and nothing else, which is why it works with no network at all.

If you only want the intelligence

You do not have to install anything. All 31 tools work with nothing but the MCP connection, and they cover every lookup — a CVE, its exploits, its KEV status, a package’s advisories. The prompts are for when you want the workflow around that intelligence, and they are never a precondition for it.