Atlas / MCP servers / tercumantanumut / Selene

SeleneBLOCK

Name: Selene
Author: tercumantanumut

mcp/tercumantanumut/selene

MCP server by tercumantanumut · agent · antigravity · codebase indexing · Audited 2026-10-06

Selene is a desktop app that runs AI agents on your machine. Connect them to your WhatsApp, Telegram, Slack, or Discord. Write code, generate images, build personal assistants. All from one place. Your data stays on your device.

Verdict

BLOCK

Grade

Trust score

34 /100

Exposed tools

25 23r · 2w · 0d

Transport

streamable-http

License

MIT

Stars

169

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

Selene is an agent-first desktop app that runs AI on your machine. Chat, write code, generate images, design UIs, control a browser — then pipe all of it into WhatsApp, Telegram, Slack, or Discord. Your data stays on your device. Every part of Selene — chat, embeddings, voice, images — lets you pick between local and cloud. Run fully offline or bring your own API keys. Mix and match.

Read from source at commit d06600fb987fOBSERVED · 2026-10-06

Connect

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. Replace the environment placeholders with a token scoped to the least it needs.

claude-code

claude mcp add app-mockup-kit --env ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY} --env ANTHROPIC_UPSTREAM_API_KEY=${ANTHROPIC_UPSTREAM_API_KEY} --env APPLE_API_KEY=${APPLE_API_KEY} --env APPLE_API_KEY_ID=${APPLE_API_KEY_ID} -- npx -y [email protected]

claude-desktop

{
  "mcpServers": {
    "app-mockup-kit": {
      "command": "npx",
      "args": [
        "-y",
        "[email protected]"
      ],
      "env": {
        "ANTHROPIC_API_KEY": "${ANTHROPIC_API_KEY}",
        "ANTHROPIC_UPSTREAM_API_KEY": "${ANTHROPIC_UPSTREAM_API_KEY}",
        "APPLE_API_KEY": "${APPLE_API_KEY}",
        "APPLE_API_KEY_ID": "${APPLE_API_KEY_ID}"
      }
    }
  }
}

Exposed tools (25)

23 read · 2 write · 0 destructive.

Tool	Risk	Description
`Coda`	write	Create and manage Coda docs, pages, and tables
`Composio`	read	Connect many apps through Composio
`Everything`	read	Sample server with many example tools
`GitHub`	read	Manage repositories and search code
`InvalidName`	read	Has uppercase
`Linear`	read	Track issues and manage projects
`PostgreSQL`	read	Connect to a PostgreSQL database
`Skill`	read	Desc
`Supabase`	read	Manage Supabase project data and APIs
`app`	read	app/, components/, hooks/, i18n/, middleware/instrumentation
`code-review`	read	Specialized agents for reviewing pull requests with code intelligence
`commit-commands`	write	Git commit workflows including commit, push, and PR creation
`electron`	read	electron/ main + preload
`folder-plugin`	read	Imported from folder
`hookify`	read	Hook management plugin
`lib`	read	lib/ and shared type declarations
`no-server`	read	missing server
`old-skill`	read	Upgraded to full plugin
`plugin-a`	read	Plugin A
`plugin-b`	read	Plugin B
`reviewer`	read	Code review specialist agent
`test-plugin`	read	A comprehensive test plugin
`tooling`	read	root config files and scripts/
`versionless-plugin`	read	Plugin without explicit version
`x`	read	Single char

Trust audit

BLOCKgrade F · trust 34/100 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	WARN
L1	Static analysis of the code	FAIL
L2	Instruction surface (what it tells the agent)	FAIL
L3	Class-specific surface	PASS
L4	Behavioural (sandbox)	SKIPPED

What the source does

Filesystem: declared (14 observation(s))
Network: declared (17 observation(s))
Shell: declared (10 observation(s))
Dependencies: not all pinned
Secrets in source: found

Findings (25)

HIGHCode injection · code.eval_exec · CWE-78, CWE-94, CWE-95

electron/metadata-collector.ts:7

exec(command, { timeout }, (error, stdout) => {

Why it matters. evaluates text as code

Fix. remove; use a parser or a dispatch table

HIGHCode injection · code.eval_exec · CWE-78, CWE-94, CWE-95

lib/ai/tools/chromium-workspace-tool.ts:383

const textContent = await session.page.$eval(input.selector, (el) => el.textContent ?? "");

Why it matters. evaluates text as code

Fix. remove; use a parser or a dispatch table

HIGHCode injection · code.eval_exec · CWE-78, CWE-94, CWE-95

lib/ai/tools/chromium-workspace-tool.ts:413

return eval(expr);

Why it matters. evaluates text as code

Fix. remove; use a parser or a dispatch table

HIGHCode injection · code.eval_exec · CWE-78, CWE-94, CWE-95

lib/ai/transformers-runtime.ts:40

const dynamicImport = new Function("specifier", "return import(specifier)") as (

Why it matters. evaluates text as code

Fix. remove; use a parser or a dispatch table

HIGHCode injection · code.eval_exec · CWE-78, CWE-94, CWE-95

lib/documents/pdf-parse-runtime.ts:40

const dynamicImport = new Function("specifier", "return import(specifier)") as (

Why it matters. evaluates text as code

Fix. remove; use a parser or a dispatch table

HIGHNetwork egress · net.metadata_ip · CWE-200, CWE-319

app/api/browser/[sessionId]/interact/route.ts:181

hostname === "169.254.169.254" ||

Why it matters. cloud metadata endpoint: the classic SSRF credential grab

HIGHNetwork egress · net.metadata_ip · CWE-200, CWE-319

app/api/browser/[sessionId]/interact/route.ts:182

hostname === "metadata.google.internal" ||

Why it matters. cloud metadata endpoint: the classic SSRF credential grab

HIGHPrompt injection · prompt.tool_poisoning · CWE-94, CWE-1427

lib/skills/catalog/bundled/security-best-practices.md:56

Important: When referencing code in the report, make sure to find and include line numbers for the code you are referencing.

Why it matters. a tool description carrying instructions to the agent

Fix. tool descriptions describe the tool; nothing else