// MCP TOOL REFERENCE

DOLLHOUSEMCP TOOLS

5 tools from the DollhouseMCP MCP Server, categorised by risk level.

READ TOOLS

mcp_aql_read Safe, read-only operations. Supported operations: activate_element, open_portfolio_browser, open_logs, open_metrics, open_permissions, open_setup,...

WRITE TOOLS

mcp_aql_create Additive, non-destructive operations. Supported operations: create_element, import_element, addEntry, verify_challenge, beetlejuice_beetlejuice_be... 2/5 mcp_aql_update Modifying operations that overwrite data. Supported operations: edit_element, upgrade_element Element types: persona, skill, template, agent, mem... 2/5

DESTRUCTIVE TOOLS

mcp_aql_delete Destructive operations that remove data. Supported operations: delete_element, clear, clear_github_auth Element types: persona, skill, template, ... 4/5

EXECUTE TOOLS

mcp_aql_execute Execution lifecycle operations for executable elements (agents, workflows, pipelines). Supported operations: confirm_operation, approve_cli_permis... 3/5

// FAQ

How many tools does the DollhouseMCP MCP server have? +

The DollhouseMCP MCP server exposes 5 tools across 4 categories: Read, Write, Destructive, Execute.

How do I enforce policies on DollhouseMCP tools? +

Use Intercept, the open-source MCP proxy. Write YAML rules for each tool — rate limits, argument validation, or deny rules — then run Intercept in front of the DollhouseMCP server.

What risk categories do DollhouseMCP tools fall into? +

DollhouseMCP tools are categorised as Read (1), Write (2), Destructive (1), Execute (1). Each category has a recommended default policy.

Enforce policies on DollhouseMCP

Open source. One binary. Zero dependencies.

npx -y @policylayer/intercept

github.com/policylayer/intercept →