SceneScoutBLOCK
🔭 Your coding agent, turned into a QA tester for any web app. Finds what is broken and shows you the proof: pictures, replays, journey videos and test reports. Over MCP, with Claude Code, Cursor, VS Code, Codex, Gemini CLI and more.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Your coding agent, turned into an exploratory QA tester for any web app.
Works with Claude Code · Cursor · VS Code (Copilot) · Codex CLI · Gemini CLI · Copilot CLI · Windsurf · any MCP client
[](https://github.com/brunoboto96/SceneScout/actions/workflows/test.yml) [](https://www.npmjs.com/package/scenescout) [](LICENSE)
🌐 Website · 📖 Guide · 🐛 What it catches · 🎬 QA with evidence · 🚀 Get started · 🚦 CI · 🔒 Safety · 📚 Docs
Scripted end-to-end tests answer one question: does this exact flow still work? They say nothing about the rest of the app. SceneScout lets the agent you already use explore a running web app like a curious, thorough tester. It clicks, fills forms, switches roles and calls the API behind a hidden button, then writes a report of what is broken and what could be better, with a picture and evidence for every line.
Try it on any app you are allowed to test. No account, no API key, no setup:
npx -y scenescout http://localhost:3000
flowchart LR A["🧠 Your coding agentClaude Code, Cursor, Copilot..."] -- MCP --> S["🔭 SceneScoutbrowser · checks · memory"] S --> App["🌐 Your running app"] S --> R["📋 Reportplain words + pictures"] S --> E["🎬 Evidencereplay, videos, test report"]
👥 Who it's for
739ea5326f2cOBSERVED · 2026-10-11Connect
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 mcp add scenescout --env EXPECTED_KEY=${EXPECTED_KEY} --env PASSED=${PASSED} -- npx -y [email protected]{
"mcpServers": {
"scenescout": {
"command": "npx",
"args": [
"-y",
"[email protected]"
],
"env": {
"EXPECTED_KEY": "${EXPECTED_KEY}",
"PASSED": "${PASSED}"
}
}
}
}Exposed tools (39)
36 read · 3 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
focus | read | An area or flow to concentrate on |
judge_pair | read | Record whether the two findings describe one defect. |
level | read | How far to go: ${LEVELS.join( |
role | write | The name to save the sign-in under, e.g. admin. scout_attach { role } uses it afterwards. Not a password. |
scout_attach | read | attach |
scout_back | read | |
scout_capture | read | |
scout_click | read | click |
scout_close | read | |
scout_coverage | read | |
scout_crawl | read | crawl |
scout_criterion | read | |
scout_design_audit | read | |
scout_finding | read | |
scout_hover | read | |
scout_journey | read | |
scout_lane_brief | read | |
scout_lane_report | read | |
scout_login | read | |
scout_navigate | read | |
scout_network | read | |
scout_new_tool | read | not yet decided |
scout_note | read | |
scout_press | read | |
scout_report | read | |
scout_request | read | |
scout_resolve | read | |
scout_run_plan | write | |
scout_scan | read | |
scout_screenshot | read | shot |
scout_scroll | read | |
scout_select | read | |
scout_session | read | |
scout_snapshot | read | |
scout_tickets | read | |
scout_type | read | |
scout_upload | write | |
scout_verify | read | |
url | read | URL of the running app, e.g. http://localhost:3000 |
Trust audit
BLOCKgrade F · trust 25/100 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 | FAIL |
| 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 (8 observation(s))
- Network
- declared (3 observation(s))
- Shell
- declared (5 observation(s))
- Dependencies
- not all pinned
- Secrets in source
- found
Findings (25)
ci-fromrun-demo-c2.json
ci-fromrun-small-demo-c2.json
const liveFactsOf = new Function("node", LIVE_FACTS_BODY) as (node: Element) => LiveFacts;description: On Linux, install the system libraries the browser needs (uses sudo). Set to false on a runner that already has them.
else sudo "$(command -v node)" "$PLAYWRIGHT_CLI" install-deps "$ENGINE"; fi
- **When the page and the server disagree, look at what the page asked: `scout_network`.** It lists the fetch/XHR requests the current page made since it loaded — method, path, status, time, and the r
- **A `postmessage_token` violation is a `security` finding at high.** The page called `postMessage` with targetOrigin `"*"` and a token-shaped value in the message (a JWT, a `Bearer` value, or an opa
`worth-a-look` output. `--ignore` takes them like any rule. Every other rule
- `--ignore` removes them like any other rule.
const asked = login ? ` Asked for by @${inert(login, 60)}.` : "";live: { port: 4321, token: "tok_abcdefghijklmnop" },password: "SCENESCOUT_LOGIN_PASSWORD",
assert.equal(secrets("-----BEGIN OPENSSH PRIVATE KEY-----"), 1);.prettierignore
.prettierrc.json
const show = new Function(`return ${shown[1]};`)() as RegExp;const drop = new Function(`return ${dropped[1]};`)() as RegExp;assert.doesNotThrow(() => new Function(script));
assert.doesNotThrow(() => new Function(script));
"secrets/id_rsa",
"home/.ssh/config",
for (const name of ["a.pdf", "a.docx", "id_rsa", "a.md.png", "a.json"]) assert.ok(!isTicketFileName(name), name);
for (const bad of ["/etc/passwd", "C:/x.md", "../outside.md", "docs/", " "]) {[{ action: "upload", path: "/etc/passwd" }, /steps\[1\] unknown field\(s\) "path"/],for (const bad of ["recordings/../x/0001-a.jpg", "/etc/passwd", "recordings/check/finding-abc.png", "recordings/a/b/0001-a.jpg", ""])
Gates applied: instruction_override, no_behavioural_pass.
739ea5326f2cfull audit observations/trust-audit/mcp-server/brunoboto96__scenescout.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 |
|---|---|---|---|---|---|
| 2026-10-11 | 739ea5326f2c | BLOCK | F | 25 | first audit |
Questions
What is the SceneScout MCP server?
🔭 Your coding agent, turned into a QA tester for any web app. Finds what is broken and shows you the proof: pictures, replays, journey videos and test reports. Over MCP, with Claude Code, Cursor, VS Code, Codex, Gemini CLI and more.
What tools does SceneScout expose?
39 in total: 36 read-only, 3 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is SceneScout safe to connect to an agent?
No — not without reading the findings first. The audit graded it F (25/100) and found 9 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 SceneScout need?
It reads EXPECTED_KEY and PASSED from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does SceneScout run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as scenescout at 3.23.4.
How current is this page?
The grade is for one exact copy of the source (739ea5326f2c), read on 2026-10-11. The repository is watched and re-audited when it changes.