stigg-mcp

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill contains command and config snippets that ask the user to insert their API key verbatim (HTTP header, JSON config, and env var), which would require an LLM to handle or emit secret values directly.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill requires connecting to and fetching a tool catalog from the MCP server at https://mcp.stigg.io (and suggests re-fetching runtime docs at https://docs.stigg.io/api-and-sdks/mcp-server), which are runtime endpoints that control the agent's available tools and behavior; additionally the skill instructs running a remote stdio bridge via npx ("@stigg/typescript-mcp") which fetches and executes remote code at runtime.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (high risk: 1.00). The skill explicitly exposes Stigg billing/metering functionality to the agent: it lets the agent provision customers, create/preview/cancel subscriptions, grant entitlements, and report usage/metering events, and instructs how to supply an API key so the MCP will forward API calls on the user's behalf. These are explicit financial/billing operations (directly affecting customer subscriptions and usage billing), so this grants direct financial execution capability.