// MCP TOOL REFERENCE

SYSTEMPROMPT CODING AGENT TOOLS

8 tools from the SystemPrompt Coding Agent MCP Server, categorised by risk level.

View the SystemPrompt Coding Agent policy →

// ALL 8 SYSTEMPROMPT CODING AGENT TOOLS

READ 3 tools

Read get_prompt Get a specific prompt by ID or list all available prompts if no ID is provided Read report Generate a report on task status. Shows all tasks if no ID provided, or details for a specific task. Read update_stats Get current statistics on tasks and active sessions

DESTRUCTIVE 2 tools

Destructive clean_state Clean up system state by removing all tasks Destructive end_task End a task and clean up the AI model session

EXECUTE 3 tools

Execute check_status Check the status of Claude Code SDK and Gemini CLI availability by attempting to initiate test sessions Execute create_task Create a new task and start it immediately with Claude Code Execute update_task Send instructions to an active AI process (Claude Code or Gemini CLI)

// FAQ

How many tools does the SystemPrompt Coding Agent MCP server have? +

The SystemPrompt Coding Agent MCP server exposes 8 tools across 3 categories: Read, Destructive, Execute.

How do I enforce policies on SystemPrompt Coding Agent tools? +

Route the SystemPrompt Coding Agent server through the PolicyLayer gateway. Define allow, deny, or approval rules per tool in the dashboard; they are enforced on every call before it reaches the server.

What risk categories do SystemPrompt Coding Agent tools fall into? +

SystemPrompt Coding Agent tools are categorised as Read (3), Destructive (2), Execute (3). Each category has a recommended default policy.

Enforce policy on every SystemPrompt Coding Agent tool call.

Deterministic rules across all 8 SystemPrompt Coding Agent tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

GOVERN SYSTEMPROMPT CODING AGENT →

Free to start. No card required.

42,500+ MCP servers and 110,000+ tools scanned and risk-classified.