Skills/Docker MCP Catalog/apify-mcp-server

apify-mcp-server@0.21.11

CAUTION· install only

Publisher claims: "Apify is the world's largest marketplace of tools for web scraping, data extraction, and web automation. You can extract structured data from social media, e-commerce, search engines, maps, travel sites, or any other website." Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry Docker MCP CatalogPublisher apify · publishing for 667 daysInstalls 39,446Last detonated 10 h ago
What would fix thisWatchDispute verdict

Coverage

Nothing executable in the bundle ran; the trace covers the install step only.
Tested
every file read, file written and host contacted by what ran
Not tested
use of the skill by an agent

What this means for the label: Caution here covers what ran. The server's tools were not exercised with a real credential, so nothing about their behavior is asserted; the credential summary below says what to do with that.

Before you give it your token

It will ask for
APIFY_TOKENConfiguration, not credentials: APIFY_API_BASE_URL, APIFY_API_PUBLIC_BASE_URL, APIFY_IS_AT_HOME, FIXTURE_KEY_VALUE_STORE_NAME, SKYFIRE_PAY_ID_KEY, SKYFIRE_SELLER_SERVICE_ID, TELEMETRY_ENABLED, TELEMETRY_ENV, UI_MODE
Where it talks to
apify.com, apifyusercontent.com (its own)Also crawlee.dev, amazonaws.com, gstatic.com, nodejs.org — not the publisher's domain; check what for before connecting.
What the honeypot saw
no credential files read · no network
What we could not test
use of the skill by an agent

Connect it with a scoped credential, under the guard, and read the third-party hosts first. The code names crawlee.dev, amazonaws.com, gstatic.com, nodejs.org as well as its own vendor. The guard injects APIFY_TOKEN only into requests to apify.com (and its subdomains) and refuses the rest.

$vet-skills run --secrets APIFY_TOKEN=apify.com -- docker run -i --rm mcp/apify-mcp-server

Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.

Is [email protected] safe to install?

KillerSkills ran it in the KillerSkills honeypot sandbox (1 scenario; nothing executable in the bundle ran; the trace covers the install step only) on 2026-11-14 and scored it 69/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Credential pass-through configuration; Collapsed <details> block carrying instructions; Collapsed <details> block carrying instructions. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 0 s · 1 scenarios · 0 deviations
t+CategoryObserved action
0.0sprocessno install scripts in bundle
0.0sprocessscript backend runs setup only
0.0sprocessscript backend runs setup only
0.0sprocessno install scripts to re-run

Static findings

MEDIUM
MCP server using network transport (SSE/HTTP)src/index.ts:10 · MCP server using network transport (SSE/HTTP) — data sent over networkexport { ActorsMcpServer, createStatelessServer };Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.
AST04
MEDIUM
MCP server using network transport (SSE/HTTP)src/index.ts:3 · MCP server using network transport (SSE/HTTP) — data sent over networkKeep this surface minimal: the `ActorsMcpServer` facade, plus `createStatelessServer` — theFix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.
AST04
HIGH
Credential pass-through configurationsrc/utils/auth.ts:33 · Credential pass-through configuration — agent relays sensitive credentials// Only bypass token if all requested tools are public AND no specific actors requested.Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST02
MEDIUM
npx invocation of non-well-known packagesrc/server_card.ts:31 · npx invocation of non-well-known package — may execute untrusted code* Undefined on a broken install (e.g. a truncated npx cache), which must not stop the stdio server.Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
MCP server using network transport (SSE/HTTP).claude/skills/creating-mcp-agent-evals/SKILL.md:18 · MCP server using network transport (SSE/HTTP) — data sent over network…you get input values the schema rejects. 3. **Two datasets, one per CI gate.** `mcp-server-evals-pr` (the `--dataset` default) holds the `kind: "tool-call"` items and gates PRs (PR open/reopen, or the `validated` label); `mcp-server-evals-…Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
AST04
MEDIUM
MCP server using network transport (SSE/HTTP)src/mcp/AGENTS.md:63 · MCP server using network transport (SSE/HTTP) — data sent over networkto `LegacyMcpServer`; `createStatelessServer` builds the stateless adapter per request from theFix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.
AST04
MEDIUM
MCP server using network transport (SSE/HTTP)src/dev_server.ts:156 · MCP server using network transport (SSE/HTTP) — data sent over network* stateful handling. Mirrors `apify-mcp-server-internal`'s `isSessionlessEraRequest`.Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.
AST04
HIGH
Collapsed <details> block carrying instructionsREADME.md:69 · Collapsed <details> block carrying instructions<summary><b>Claude Code</b></summary> ```bash claude mcp add apify "https://mcp.apify.com" -t http ``` Claude Code opens the Apify sign-in page the first time iFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
HIGH
Collapsed <details> block carrying instructionsREADME.md:108 · Collapsed <details> block carrying instructions<summary><b>VS Code and GitHub Copilot</b></summary> Use the one-click install link from the [configurator](https://mcp.apify.com), run `apify mcp install vscodFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
npx invocation of non-well-known packagesrc/stdio.ts:105 · npx invocation of non-well-known package — may execute untrusted code'To connect, set your MCP client server command to `npx @apify/actors-mcp-server`' +Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
HIGH
Access to sensitive file pathssrc/stdio.ts:58 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files.env()Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
MCP server using network transport (SSE/HTTP)src/tools/AGENTS.md:141 · MCP server using network transport (SSE/HTTP) — data sent over network…CaseField` in `utils/tools_loader.ts` removes it from the served schema; `ActorsMcpServer.composeToolsForClient` applies that for clients `isStoreUseCaseServable` does not allow. The call passes it to `searchAgentSafeActors`, which sanitiz…Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.
AST04
HIGH
Collapsed <details> block carrying instructionsREADME.md:80 · Collapsed <details> block carrying instructions<summary><b>ChatGPT</b></summary> 1. Open [ChatGPT Plugins](https://chatgpt.com/plugins), select **+**, then **Add custom MCP server**. 2. Name it `Apify`, enteFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
File type combined with upload actionsrc/web/AGENTS.md:7 · File type combined with upload action — agent may exfiltrate local data files`package.json`, Tailwind/PostCSS config, and build** — the root `tsconfig.json`Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
MCP server using network transport (SSE/HTTP)src/AGENTS.md:12 · MCP server using network transport (SSE/HTTP) — data sent over network- `index.ts` — the library export (`ActorsMcpServer` plus `createStatelessServer`, theFix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.
AST04
HIGH
URL with embedded credentialssrc/mcp/actors.ts:65 · URL with embedded credentials — may be phishing or credential leaking// `origin` excludes userinfo, so `https://user:pass@<standby host>/mcp` passes the origin check; drop it here.Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST07
MEDIUM
npx invocation of non-well-known package.claude/skills/bug-triage/SKILL.md:88 · npx invocation of non-well-known package — may execute untrusted code4. **npx cache corruption**: Stale cache prevents stdio server from starting. Fix: `rm -rf ~/.npm/_npx` (macOS/Linux) or `rmdir /s /q %LOCALAPPDATA%\npm-cache\_npx` (Windows).Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known package.claude/skills/bug-triage/SKILL.md:86 · npx invocation of non-well-known package — may execute untrusted code…server"**: Claude Desktop extension install issues. Usually caused by corrupted npx cache, Claude silently downgrading the extension, or Node.js not on the GUI app's PATH…Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
HIGH
Access to sensitive file pathssrc/mcp/server.ts:246 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filestelemetryEnv = getTelemetryEnv(this.options.telemetry?.env ?? process.env.TELEMETRY_ENV);Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
MCP server using network transport (SSE/HTTP).claude/skills/dig/SKILL.md:149 · MCP server using network transport (SSE/HTTP) — data sent over network| **Internal repo** | `../apify-mcp-server-internal` (if available) | Hosted server — assess impact of changes |Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.
AST04
HIGH
Collapsed <details> block carrying instructionsREADME.md:404 · Collapsed <details> block carrying instructions<summary><b>How the server adapts Actor input schemas</b></summary> The server adapts Actor input schemas so that more clients and models can use them: - DescriFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
HIGH
Collapsed <details> block carrying instructionsREADME.md:150 · Collapsed <details> block carrying instructions<summary><b>Local stdio (npx)</b></summary> Requires Node.js 22 or later. ```json { "mcpServers": { "actors-mcp-server": { "command": "npx", "args": ["-y", "@apFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
HIGH
Collapsed <details> block carrying instructionsREADME.md:95 · Collapsed <details> block carrying instructions<summary><b>Cursor</b></summary> Use the one-click install link from the [configurator](https://mcp.apify.com), or let the [Apify CLI](https://docs.apify.com/clFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01

Cost to the agent

Instructions loaded
≈15,892 tok
8 files · 80 in bundle

Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.

Known vulnerabilities

Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.

What would make this Verified

11 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%
  1. Unclaimed publisher≈ +8 on the totalClaim apify: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
  2. Hidden instructions≈ +6 on the totalRemove HTML comments, zero-width characters and white-on-white text that carry instructions. Everything the agent is meant to read should be visible to the person reading the file.
  3. Credential pass-through configuration≈ +4 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
  4. Access to sensitive file paths≈ +4 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
  5. URL with embedded credentials≈ +4 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
  6. MCP server using network transport (SSE/HTTP)≈ +2 on the totalNamespace tool names and keep tool descriptions literal; no instructions to the agent inside them.
Publish a new version to Docker MCP Catalog, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim apifyPropose a patch to the maintainer