Skip to content
verify mcp Beta VerifyMCP is currently in beta. If you notice any issues, get in touch and we’ll put it right.

AgentDrive

REMOTE · MCP.AGENTDRIVE.SH · SCANNED SEP 20

Cross-agent artifact workspace with provenance across Claude Code, Codex, Cursor, LangGraph.

0 this week 73 Trust /100
Trust breakdown (7 categories)

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. How we score → Why this is hard to score →

Endpoint Security46
Transport & Reachability100
Schema Quality & AI Usability77
  • AI-judged instruction clarity (excellent).Pass
  • Tool/resource definitions use about 1805 tokens (~90/item across 20 items; 20 tools + 0 resources), lean.Pass
  • Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management100
  • No destabilizing schema changes in the last 30 days.Pass
Tool Coverage99
  • 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
  • 98% of tool parameters carry a description.Partial
Tool Safety75
  • No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
  • 0 of 2 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "drive_delete" implies "delete" and declares no destructiveHint at all, which the MCP spec reads as destructive by default. See how to fix → Fail
  • An AI judge read all 21 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
  • Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
Install

How do I install the AgentDrive MCP server?

AgentDrive is a hosted endpoint at https://mcp.agentdrive.sh/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

remote · mcp.agentdrive.sh

# add to Claude Code
claude mcp add --transport http sh-agentdrive-agentdrive 'https://mcp.agentdrive.sh/mcp'
// .cursor/mcp.json
{
  "mcpServers": {
    "sh-agentdrive-agentdrive": {
      "url": "https://mcp.agentdrive.sh/mcp"
    }
  }
}
// .vscode/mcp.json
{
  "servers": {
    "sh-agentdrive-agentdrive": {
      "type": "http",
      "url": "https://mcp.agentdrive.sh/mcp"
    }
  }
}
# ~/.codex/config.toml
[mcp_servers.sh-agentdrive-agentdrive]
url = "https://mcp.agentdrive.sh/mcp"
// opencode.json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "sh-agentdrive-agentdrive": {
      "type": "remote",
      "url": "https://mcp.agentdrive.sh/mcp",
      "enabled": true
    }
  }
}
# add to OpenClaw
openclaw mcp add sh-agentdrive-agentdrive --url 'https://mcp.agentdrive.sh/mcp' --transport streamable-http
# ~/.hermes/config.yaml
mcp_servers:
  sh-agentdrive-agentdrive:
    url: "https://mcp.agentdrive.sh/mcp"
// ~/.netclaw/config/netclaw.json
{
  "McpServers": {
    "sh-agentdrive-agentdrive": {
      "Transport": "http",
      "Url": "https://mcp.agentdrive.sh/mcp"
    }
  }
}
# add to Vellum
assistant mcp add sh-agentdrive-agentdrive -t streamable-http -u 'https://mcp.agentdrive.sh/mcp'
// mcp.json
{
  "mcpServers": {
    "sh-agentdrive-agentdrive": {
      "type": "http",
      "url": "https://mcp.agentdrive.sh/mcp"
    }
  }
}

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

Changelog

Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.

  • 26 Aug 26 −1
    • We updated how we score, so this day's move reflects our rubric, not a change to the server See what changed → functional
  • 25 Aug 26 +1
    • Stability: 0.97 → pass security
  • 23 Aug 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 90 to 93. That category is still filling its 30-day observation window: 27 days of observed history at the previous scan, 28 at this one. The score rises as the window fills, whether or not the server changes.

  • 11 Aug 26 0
    • We updated how we score, so this day's move reflects our rubric, not a change to the server See what changed → functional
  • 31 Jul 26 0
    • We updated how we score, so this day's move reflects our rubric, not a change to the server See what changed → functional
  • 27 Jul 26 0
    • We updated how we score, so this day's move reflects our rubric, not a change to the server See what changed → functional
  • 26 Jul 26 0

    First indexed and scored.

Diagnostics

Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.

Captured 20 Sept 2026 · Probed https://mcp.agentdrive.sh/mcp

TLS valid

Negotiated TLS 1.3 with TLS_AES_128_GCM_SHA256 .

Subject Issuer Valid from Valid until Key Signature Serial
CN=agentdrive.sh CN=WE1,O=Google Trust Services,C=US 1 Aug 2026 30 Oct 2026 ECDSA 256 ECDSA-SHA256 48b7913b760173e40eed63940d418462
SANs: agentdrive.sh, mcp.agentdrive.sh, *.mcp.agentdrive.sh
CN=WE1,O=Google Trust Services,C=US (CA) CN=GTS Root R4,O=Google Trust Services LLC,C=US 13 Dec 2023 20 Feb 2029 ECDSA 256 ECDSA-SHA384 7ff31977972c224a76155d13b6d685e3
CN=GTS Root R4,O=Google Trust Services LLC,C=US (CA) CN=GlobalSign Root CA,OU=Root CA,O=GlobalSign nv-sa,C=BE 15 Nov 2023 28 Jan 2028 ECDSA 384 SHA256-RSA 7fe530bf331343bedd821610493d8a1b

Background: What to check on a remote MCP endpoint →

DNSSEC insecure

Validation of mcp.agentdrive.sh. Not signed

Zone DS Keys Algorithms Outcome
. trust_anchor 20326, 38696 8, 8 Verified
sh. present 55297 8 Verified
agentdrive.sh. absent Unsigned (proven) parent-signed NSEC/NSEC3 proves an unsigned delegation
Authentication No authorisation required

The endpoint answered without asking for a token. Anyone who knows the URL can reach it.

Result No authorisation required
HTTP status 200

Background: How OAuth 2.1 works in the 2026 MCP spec →

Transports 2 probes
Transport URL Outcome Status Location
streamable-http https://mcp.agentdrive.sh/mcp Verified 200
http (plaintext) http://mcp.agentdrive.sh/mcp Inconclusive 405
MCP tools · 20 exposed · ~1,602 tokens

The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →

Tool Tokens
drive_activity ~32

Get recent activity in a workspace (file writes, comments, shares).

NameTypeReqDescription
workspace_idstringyesWorkspace ID

No output schema declared.

No examples provided.

drive_comment ~73

Add a comment to a file.

NameTypeReqDescription
bodystringyesComment text
file_idstringyesFile ID to comment on
line_endnumberEnd line number (optional)
line_startnumberStart line number (optional)
workspace_idstringyesWorkspace ID

No output schema declared.

No examples provided.

drive_context ~124

Auto-pull the most relevant artifacts from prior agent sessions for the task you are about to do. Combines semantic relevance with recency. Call this at the START of any non-trivial task to ground the agent in prior context — replaces manual copy-paste of artifacts between sessions.

NameTypeReqDescription
limitintegerMax artifacts to return (default 5, max 20)
taskstringyesWhat you are about to do — used to rank prior artifacts by relevance
workspace_idstringScope to a single workspace; omit for org-wide context

No output schema declared.

No examples provided.

drive_delete ~37

Delete a file from a workspace.

NameTypeReqDescription
file_idstringyesFile ID to delete
workspace_idstringyesWorkspace ID

No output schema declared.

No examples provided.

drive_feedback ~49

Read human comments on a file. Use this to check if humans have left feedback.

NameTypeReqDescription
file_idstringyesFile ID to read comments from
workspace_idstringyesWorkspace ID

No output schema declared.

No examples provided.

drive_invite ~59

Invite a human to your organization by email. They will receive a verification code to join.

NameTypeReqDescription
emailstringyesEmail address of the human to invite to your organization
rolestringRole in the organization (default: member)

No output schema declared.

No examples provided.

drive_list ~46

List files in a workspace, optionally filtered by path prefix.

NameTypeReqDescription
prefixstringPath prefix filter (e.g. "notes/")
workspace_idstringyesWorkspace ID

No output schema declared.

No examples provided.

drive_log_run ~156

Log an agent run to AgentDrive's provenance store. Captures task, status, referenced/produced artifacts, and cross-platform identity (which agent on which tool). Used for retrospective context retrieval and team-wide audit.

NameTypeReqDescription
artifacts_producedarrayPaths of artifacts the run wrote
artifacts_referencedarrayPaths of artifacts the run read
duration_msintegerWall-clock duration in milliseconds
metadataobjectFree-form key/value metadata
run_idstringExisting run ID to update; omit to let server allocate
statusstringRun outcome (default success)
taskstringyesWhat the agent did or is doing (free text)

No output schema declared.

No examples provided.

drive_org ~22

View your organization details: agents, members, plan, and usage.

Input schema present but exposes no named parameters.

No output schema declared.

No examples provided.

drive_read ~39

Read a file's content from a workspace.

NameTypeReqDescription
file_idstringyesFile ID to read
workspace_idstringyesWorkspace ID

No output schema declared.

No examples provided.

drive_recall ~89

Recall persistent facts and preferences your agents have written. In v1 this returns semantic search results across artifacts; v1.5 will narrow to auto-extracted facts (preferences, decisions, key entities).

NameTypeReqDescription
querystringyesRecall query — what fact, preference, or decision are you trying to remember?
top_kintegerMax results (default 5, max 20)

No output schema declared.

No examples provided.

drive_roles ~88

Manage per-workspace role delegation. Assign, list, or remove access for specific emails (view/comment/edit).

NameTypeReqDescription
actionstringyesAction: assign a role, list all roles, or remove a role
emailstringEmail address (required for assign/remove)
permissionstringPermission level (required for assign)
workspace_idstringyesWorkspace ID

No output schema declared.

No examples provided.

drive_sandbox_clone ~118

[EXPERIMENTAL] Clone a git repo into a per-workspace Linux sandbox. Requires Cloudflare Containers on the AgentDrive worker; gracefully degrades to a 503 with "sandbox not configured" otherwise.

NameTypeReqDescription
branchstringBranch to check out (default: repo default)
depthintegerShallow clone depth (default: full clone)
urlstringyesGit repository URL to clone
workspace_idstringyesWorkspace to clone into. The sandbox persists across MCP sessions per workspace.

No output schema declared.

No examples provided.

drive_sandbox_commit ~107

[EXPERIMENTAL] Stage + commit the current sandbox worktree. Surfaces as an AgentDrive artifact at commits/<sha>.json. Requires Cloudflare Containers; degrades to 503 otherwise.

NameTypeReqDescription
author_emailstringOverride git author email (default: agent identity)
author_namestringOverride git author name (default: agent identity)
messagestringyesCommit message
workspace_idstringyesWorkspace whose sandbox to commit

No output schema declared.

No examples provided.

drive_sandbox_exec ~124

[EXPERIMENTAL] Run a shell command inside the per-workspace Linux sandbox. Captures stdout/stderr (64 KB cap), exit code, duration. Requires Cloudflare Containers; degrades to 503 otherwise.

NameTypeReqDescription
cmdstringyesShell command to run inside the per-workspace Linux sandbox
cwdstringWorking directory (default: workspace root)
timeout_msintegerMax runtime in ms (default 30000, hard cap 60000)
workspace_idstringyesWorkspace whose sandbox to run in

No output schema declared.

No examples provided.

drive_sandbox_read ~84

[EXPERIMENTAL] Read a file from the per-workspace sandbox filesystem (build artifacts, test output, logs). Requires Cloudflare Containers; degrades to 503 otherwise.

NameTypeReqDescription
pathstringyesPath inside the sandbox (e.g., src/index.ts or /tmp/build.log)
workspace_idstringyesWorkspace whose sandbox to read from

No output schema declared.

No examples provided.

drive_search ~96

Semantic search across your org's artifacts in AgentDrive. Returns the top-N most relevant files. Use this BEFORE writing similar artifacts to avoid duplication and surface prior context.

NameTypeReqDescription
filterobject
querystringyesSearch query
top_kintegerMax hits to return (default 10, max 50)
workspace_idstringScope to a single workspace; omit for org-wide search

No output schema declared.

No examples provided.

drive_share ~143

Generate a share link for a workspace or specific file. Optionally email-gate the link so only a specific person can access it. Humans open this link to view, comment, or edit files.

NameTypeReqDescription
emailstringEmail-gate: only this email can access the link (requires email verification). Leave empty for public access.
expires_innumberExpiry in seconds (optional)
file_idstringScope to a specific file (optional)
max_usesnumberMaximum number of uses (optional)
permissionstringyesPermission level for the share link
workspace_idstringyesWorkspace ID

No output schema declared.

No examples provided.

drive_versions ~38

View version history for a file in a workspace.

NameTypeReqDescription
file_idstringyesFile ID
workspace_idstringyesWorkspace ID

No output schema declared.

No examples provided.

drive_write ~78

Write a markdown file to a workspace. Creates or updates the file with automatic versioning.

NameTypeReqDescription
contentstringyesFile content (markdown)
messagestringOptional version message
pathstringyesFile path (e.g. "notes/meeting.md")
workspace_idstringyesWorkspace ID to write to

No output schema declared.

No examples provided.

Common questions

What is the AgentDrive MCP server?

AgentDrive is an MCP server listed in the public MCP registry as sh.agentdrive/agentdrive. Cross-agent artifact workspace with provenance across Claude Code, Codex, Cursor, LangGraph. This page covers its hosted endpoint (https://mcp.agentdrive.sh/mcp).

Is the AgentDrive MCP server safe to use?

AgentDrive scores 73 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

What tools does the AgentDrive MCP server expose?

AgentDrive exposes 20 tools: drive_write, drive_read, drive_list, drive_share, drive_comment, and 15 more. Their descriptions and schemas cost roughly 1,602 tokens of context every time the server is loaded.

Does the AgentDrive MCP server require authentication?

No. We connected to AgentDrive without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

Is the AgentDrive MCP server still maintained?

AgentDrive is still listed as active in the MCP registry. We last reached this channel on 20 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.