Skip to content
MCP server

Spellwarden

Scan agent skills and MCP servers, read findings and export policy packs from Claude

Pricing
Free trial
Platforms
API
Runs
Remotely
Version
1f6ff3b14430
Listing updated
2 days ago

About Spellwarden

Spellwarden is an MCP server for security. It runs as a hosted remote server, so there is nothing to install locally.

Spellwarden's own connector reaches one organisation, chosen on the consent screen, with Read only (recommended) or Read and write access limited by your role. It uses OAuth 2.1 with PKCE; access tokens last an hour, refresh tokens 30 days, and a connection lasts at most 365 days. Up to five connected apps are allowed at a time, and no tool returns an API key's value.

Read tools list organisations, skills, findings, API keys, policy packs, MCP servers and runtime inventory built from OpenTelemetry traces. Write tools start a skill scan, add an MCP server by URL, repository or registry name, export policy packs for Claude Code hooks, Kong or Zuplo, and start eval runs on suites such as Infra-Ops 40 or AppSec 40.

The private registry is Enterprise-only and eval runs are counted per month. The connector is on Team and Enterprise plans, and new organisations start with a 14-day Team trial.

Install

Run in a terminal

claude mcp add --transport http spellwarden https://dash.spellwarden.ai/api/v1/mcp

Add to Settings → Connectors → Add custom connector

https://dash.spellwarden.ai/api/v1/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "spellwarden": {
      "url": "https://dash.spellwarden.ai/api/v1/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "spellwarden": {
      "type": "http",
      "url": "https://dash.spellwarden.ai/api/v1/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "spellwarden": {
      "serverUrl": "https://dash.spellwarden.ai/api/v1/mcp"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "spellwarden": {
      "type": "http",
      "url": "https://dash.spellwarden.ai/api/v1/mcp"
    }
  }
}

Tools this server provides (19)

Taken from the project documentation; the running server may differ.

list_organisations
Lists reachable organisations and the connection's grant.
get_organisation
Shows one organisation's plan, billing and access.
list_api_keys
Lists API keys without their values.
list_skills
Lists scanned skills with scores and findings.
get_skill
Shows a skill and its latest scan findings.
scan_skill
Starts a scan of a skill in a public repository.
list_findings
Lists findings across skills, servers and repositories.
add_mcp_server
Adds an MCP server and starts its first scan.
get_migration_report
Shows an MCP server's migration report.
list_registry_entries
Lists the private registry (Enterprise).
list_policy_packs
Lists policy packs.
export_policy_pack
Exports a policy pack version with a 15-minute link.
get_trust_score
Shows the trust score and its parts.
start_eval_run
Starts an eval run in the background.
get_eval_run
Shows an eval run's progress.
export_eval
Exports eval rubrics as files.
get_runtime_inventory
Lists what agents used at runtime from OpenTelemetry traces.
list_mcp_servers
Lists the organisation's MCP servers.
list_agent_space_connections
Lists connected AWS DevOps Agent Spaces.

What it does

Skill scanning
Starts scans of skills in public GitHub or GitLab repositories.
Findings and trust score
Lists findings by severity and shows the organisation's trust score.
MCP server checks
Adds servers and returns a migration report with a checklist.
Policy pack export
Exports packs for Claude Code hooks, Kong or Zuplo.
Eval runs
Runs task suites with and without chosen skills.
Read-only default
Write tools need explicit Read and write approval.

Good for

  • Scan a new skill before enabling it
  • Review open findings across skills and servers
  • Export a policy pack for gateway enforcement
  • Audit what agents used at runtime

Pricing

Free trial

14-day Team trial; connector on Team and Enterprise plans

Prices change. Check spellwarden.ai before you buy.

Spellwarden FAQ

Which plans include it?

Team and Enterprise; new organisations get a 14-day Team trial.

Can Claude change things?

Only if you chose Read and write and your role allows it.

How long do tokens last?

Access tokens one hour, refresh tokens 30 days.

Are eval runs limited?

Yes, they count against a monthly limit for the organisation.

Is Spellwarden free?

Spellwarden is a paid product with a free trial.

How do I add Spellwarden to Claude Code?

Run this command in your terminal: claude mcp add --transport http spellwarden https://dash.spellwarden.ai/api/v1/mcp

What are the best alternatives to Spellwarden?

The closest alternatives to Spellwarden are REA, Claude Code Ultimate Guide, Agent Roast, Midplane and Leash.

Alternatives to Spellwarden

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. MCP server for searching the Claude Code Ultimate Guide from any MCP client

    Runs locally
    • Open source
    • 6.5K GitHub stars
    • AI models and LLM tooling
  3. Agent Roast

    MCP server

    Offline review that flags agent changes which weaken tests, assertions or CI

    Runs locally
    • Open source
    • 4 Hacker News points
    • Developer tools
  4. Midplane

    MCP server

    Postgres gateway for AI agents with table grants, column masking and approval for writes

    Runs locally
    • Open source
    • 5 GitHub stars
    • Databases
  5. Leash

    MCP server

    Review what your AI touched, approve risky actions and undo supported changes

    • 1 tool
    • Remote
    • Free
    • 1 GitHub star
    • Security