FastlyBLOCK
Model Context Protocol (MCP) server for AI-powered Fastly CDN management.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
The Fastly MCP server lets MCP clients work with the Fastly API.
Connect it to an assistant such as Claude Desktop, Claude Code, Gemini CLI, Opencode, Qwen Code, Cline, or Swival, and give it a Fastly API token. The assistant can then:
- Look up services and inspect domains and TLS settings.
- Check traffic and usage.
- Manage dictionaries and ACLs.
- Purge content and make configuration changes when you ask it to.
This project is actively developed and has Tier 1 Fastly open-source support. The support policy is described in Fastly's open-source documentation.
Questions and feedback are welcome on the Fastly developer tools community forum.
Contents
- How it works
- Requirements
- Installation
- Running over HTTP
- Running a remote service for several users
- Using the server well
- Secret encryption
- Troubleshooting
- Local development
- Security
- License
How it works
Most MCP servers expose one tool for each operation. Because Fastly's API is large, this server uses a smaller set of tools that an assistant can use together.
- Start with
searchto find API methods by keyword, method name, API class, or HTTP path.
- Then use
inspectto get a method's documentation, including its parameters, return type, and an example.
- Once you have the details, use
executeto run a short JavaScript snippet with theFastlyclient already set up with your API token.
This keeps the tool list small while still covering every method documented in the bundled Fastly client docs.
Requirements
Runtime
Install either [Bun](https://b
4bbb02b5150cOBSERVED · 2026-10-08Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.
claude mcp add mcp -- npx -y @fastly/[email protected]
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@fastly/[email protected]"
]
}
}
}Exposed tools (3)
2 read · 1 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
execute | write | |
inspect | read | |
search | read |
Trust audit
BLOCKgrade F · trust 46/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | FAIL |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (2 observation(s))
- Network
- declared (6 observation(s))
- Shell
- declared (3 observation(s))
- Dependencies
- not all pinned
- Secrets in source
- found
Findings (25)
IPAddressDeny=169.254.169.254 fd00:ec2::254
define METADATA_V4 = { 169.254.169.254 }console.log(`Secret shield benchmark under ${runtime}\n`);CMD ["bun", "--eval", "const r = await fetch('http://127.0.0.1:8231/healthz'); process.exit(r.ok ? 0 : 1)"]`Remote mode needs the Yama ptrace restriction, but ${YAMA_PATH} cannot be read. ` +"sk-ant-api03-q8ZrT2mXw7LpK4vN9cYb3HsJ6dFg1RtU5eWo0iPa-_Qz8XyV2nMb7CkL4jHg9DfS3aWe6RtY1uIo5pZx0cVb";
const token = "synthetic-token-restricted";
export const GITHUB_PAT = "ghp_ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghij";
const GITHUB_PAT = "ghp_ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghij";
export const UPSTREAM_SECRET = "ghp_ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghij";
const GITHUB_PAT = "ghp_ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghij";
const GITHUB_PAT = "ghp_ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghij";
tls_private_key: {"data":{"type":"tls_private_key","attributes":{"key":"-----BEGIN PRIVATE KEY-----\n...\n-----END PRIVATE KEY-----\n","name":"My private key"}}},const SLACK_BOT = "xoxb-12345678-12345678-siXA";
slack: "xoxb-1234567890-9876543210-AbCdEfGhIjKlMnOpQrStUvWx",
const STRIPE_LIVE = "sk_live_F9ZJz11XNvB3i0RhazAEAbGJdznewI9VguEStDdFtyHK";
streamable-http
import { resolveExecutionRuntime } from "../../src/execution-runtime.js";import { runCli } from "../../src/main.js";import { execute } from "../../src/tools/execute.js";import { SANDBOX_PATH } from "../../src/tools/execute.js";await import("../../src/sandbox.js");The supplied rules cover `169.254.169.254` and `fd00:ec2::254`.
docker compose exec fastly-mcp bun -e 'await fetch("http://169.254.169.254/", { signal: AbortSignal.timeout(3000) }).then(() => console.log("REACHABLE"), (e) => console.log("blocked", e.code ?? e.namesudo -u fastly-mcp curl -v --max-time 3 http://169.254.169.254/
Gates applied: no_behavioural_pass.
4bbb02b5150cfull audit observations/trust-audit/mcp-server/fastly__fastly-1.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 |
|---|---|---|---|---|---|
| 2026-10-08 | 4bbb02b5150c | BLOCK | F | 46 | first audit |
Questions
What is the Fastly MCP server?
Model Context Protocol (MCP) server for AI-powered Fastly CDN management.
What tools does Fastly expose?
3 in total: 2 read-only, 1 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is Fastly safe to connect to an agent?
No — not without reading the findings first. The audit graded it F (46/100) and found 2 critical or high issues in the source. Each one is listed on this page with the file and line it is on.
What credentials does Fastly need?
No credential environment variables were found in its source, so it appears to need none.
How does Fastly run?
It speaks streamable-http, so it runs as a service you connect to over the network. It is published on npm as @fastly/mcp at 2.3.0.
How current is this page?
The grade is for one exact copy of the source (4bbb02b5150c), read on 2026-10-08. The repository is watched and re-audited when it changes.