// MCP TOOL REFERENCE

AGENTDRIVE TOOLS

20 tools from the AgentDrive MCP Server, categorised by risk level.

View the AgentDrive policy → Token cost: 2,482 tokens →

// ALL 20 AGENTDRIVE TOOLS

READ 10 tools

Read drive_activity Get recent activity in a workspace (file writes, comments, shares). Read drive_feedback Read human comments on a file. Use this to check if humans have left feedback. Read drive_list List files in a workspace, optionally filtered by path prefix. Read drive_org View your organization details: agents, members, plan, and usage. Read drive_read Read a file's content from a workspace. Read drive_recall Recall persistent facts and preferences your agents have written. In v1 this returns semantic search result... Read drive_sandbox_clone [EXPERIMENTAL] Clone a git repo into a per-workspace Linux sandbox. Requires Cloudflare Containers on the A... Read drive_sandbox_read [EXPERIMENTAL] Read a file from the per-workspace sandbox filesystem (build artifacts, test output, logs). ... Read drive_search Semantic search across your org's artifacts in AgentDrive. Returns the top-N most relevant files. Use this ... Read drive_versions View version history for a file in a workspace.

WRITE 5 tools

Write drive_comment Add a comment to a file. Write drive_invite Invite a human to your organization by email. They will receive a verification code to join. Write drive_sandbox_commit [EXPERIMENTAL] Stage + commit the current sandbox worktree. Surfaces as an AgentDrive artifact at commits/<... Write drive_share Generate a share link for a workspace or specific file. Optionally email-gate the link so only a specific p... Write drive_write Write a markdown file to a workspace. Creates or updates the file with automatic versioning.

DESTRUCTIVE 2 tools

Destructive drive_delete Delete a file from a workspace. Destructive drive_roles Manage per-workspace role delegation. Assign, list, or remove access for specific emails (view/comment/edit).

EXECUTE 3 tools

Execute drive_context Auto-pull the most relevant artifacts from prior agent sessions for the task you are about to do. Combines ... Execute drive_log_run Log an agent run to AgentDrive's provenance store. Captures task, status, referenced/produced artifacts, an... Execute drive_sandbox_exec [EXPERIMENTAL] Run a shell command inside the per-workspace Linux sandbox. Captures stdout/stderr (64 KB ca...

// FAQ

How many tools does the AgentDrive MCP server have? +

The AgentDrive MCP server exposes 20 tools across 4 categories: Read, Write, Destructive, Execute.

How do I enforce policies on AgentDrive tools? +

Route the AgentDrive 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 AgentDrive tools fall into? +

AgentDrive tools are categorised as Read (10), Write (5), Destructive (2), Execute (3). Each category has a recommended default policy.

Let agents act without letting them run wild.

Route your MCP servers through PolicyLayer and every tool call is checked against your policy before it runs — allow, deny, or require approval. Per-identity grants. Full audit log. Live in minutes.

SECURE YOUR MCP →

Free to start. No card required.

4,600+ MCP servers and 31,000+ tools scanned and risk-classified.