AdrailsSAFE
Remote MCP server for Adrails: AI agents for Meta Ads and Google Ads that read your accounts and prepare every change for one click in Adrails.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Remote MCP server for Adrails, the AI workspace for paid media. Connect Claude, Claude Code, ChatGPT, Codex or any remote MCP client to your Meta Ads and Google Ads accounts, your store and your analytics. Agents read your accounts and business data, build campaigns and prepare budget, status and launch changes, each with a link to its exact before and after.
This repository holds the listing metadata. The server itself is hosted by Adrails: there is nothing to install or run.
Connect
You need an Adrails account. Every client uses the same URL; the client registers itself and you sign in to Adrails, then choose the workspace and the permissions on the consent screen.
Claude (claude.ai and desktop): Settings, Connectors, Add custom connector. Name it Adrails, paste https://adrails.ai/api/mcp, leave the OAuth client ID and secret empty, then connect.
Claude Code
claude mcp add --transport http adrails https://adrails.ai/api/mcp
Then run /mcp, select adrails and authenticate.
ChatGPT: turn on Developer mode in Settings (it depends on your plan and workspace policy), open Plugins and add a custom MCP server named Adrails with the URL above and OAuth authentication.
Codex
codex mcp add adrails --url https://adrails.ai/api/mcp codex mcp login adrails
Full steps, including other clients: Connect an AI client.
What it can do
63 tools, filtered by the permissions you grant and your role in the workspace:
- Performance and analysis: period and daily performanc
f7f5086784cdOBSERVED · 2026-10-05Trust audit
SAFEgrade B · trust 88/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | NA |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (3)
streamable-http
Gates applied: no_behavioural_pass, no_license.
f7f5086784cdfull audit observations/trust-audit/mcp-server/devthomy__adrails.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-05 | f7f5086784cd | SAFE | B | 88 | first audit |
Questions
What is the Adrails MCP server?
Remote MCP server for Adrails: AI agents for Meta Ads and Google Ads that read your accounts and prepare every change for one click in Adrails.
Is Adrails safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (88/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does Adrails need?
No credential environment variables were found in its source, so it appears to need none.
How does Adrails run?
It speaks streamable-http, so it runs as a service you connect to over the network.
How current is this page?
The grade is for one exact copy of the source (f7f5086784cd), read on 2026-10-05. The repository is watched and re-audited when it changes.