
FlyTrap
Mobile app QA results for coding agents: bugs with evidence, SDKs, hosts and build diffs
- Pricing
- Paid
- Platforms
- API, CLI
- Runs
- Locally
- Version
- 0.1.4
- Listing updated
- 3 days ago
About FlyTrap
FlyTrap is a paid MCP server for developer tools. It runs locally on your machine and is started by your AI client.
FlyTrap itself exercises an Android or iOS app on real devices like a new user would, and the site says it returns a list of where the app breaks in about 30 minutes. This server only reads those results: it does not run apps, decompile binaries or touch your source tree. Each bug comes with a scenario, the failed step, screenshots and UI XML, the logcat error window, failed network calls, and a ready-made prompt for the agent.
Beyond bugs it exposes the screen graph, the third-party hosts a build talks to, detected SDKs with confidence and evidence, declared permissions and deep links, and a comparison of what changed between two builds. A public SDK catalog works without a key.
Two connection modes exist. The hosted endpoint at api.flytrap.dev uses OAuth: read access is always on, and a separate checkbox lets the agent queue new crawls. The local npm package takes a FLYTRAP_API_KEY. The connection can be revoked under the dashboard's API key settings.
Install
Run in a terminal
claude mcp add mcp --env FLYTRAP_API_KEY='<FLYTRAP_API_KEY>' -- npx -y @flytrap-qa/mcpAdd to claude_desktop_config.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@flytrap-qa/mcp"
],
"env": {
"FLYTRAP_API_KEY": "<FLYTRAP_API_KEY>"
}
}
}
}Add to ~/.cursor/mcp.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@flytrap-qa/mcp"
],
"env": {
"FLYTRAP_API_KEY": "<FLYTRAP_API_KEY>"
}
}
}
}Add to .vscode/mcp.json
{
"servers": {
"mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@flytrap-qa/mcp"
],
"env": {
"FLYTRAP_API_KEY": "<FLYTRAP_API_KEY>"
}
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@flytrap-qa/mcp"
],
"env": {
"FLYTRAP_API_KEY": "<FLYTRAP_API_KEY>"
}
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@flytrap-qa/mcp"
],
"env": {
"FLYTRAP_API_KEY": "<FLYTRAP_API_KEY>"
}
}
}
}Configuration
FLYTRAP_API_KEYFlyTrap API key (ft_live_...) from https://dashboard.flytrap.dev/settings/api-keys. Optional: public catalog tools work without it.
Tools this server provides (12)
Taken from the project documentation; the running server may differ.
- whoami
- The account the key belongs to.
- list_projects
- Projects the key owner can read, with the selected crawl run.
- get_project
- One project plus its live crawl status.
- list_builds
- Builds of a project with binary SHA-256 and provenance.
- get_build_summary
- Version, size, SDK and permission counts, bug and test counts, screens and totals.
- get_sdks
- Detected SDKs with confidence and evidence, permissions and deep links.
- list_bugs
- Bugs from the selected run, sorted by severity.
- get_bug_evidence
- Scenario, failed step, screenshots, XML, logcat window and failed network calls for a bug.
- get_network_hosts
- Hosts the build talks to.
- lookup_sdk
- Looks up an SDK and its owner in the public catalog.
- compare_builds
- Differences between two builds.
- get_screen_graph
- Screens reached in the crawl.
What it does
- Bugs with evidence
- Severity-sorted bugs with screenshots, UI XML, logcat and failed network calls.
- Network host report
- See which hosts a build contacts, including before a consent screen.
- SDK and permission detection
- Detected SDKs with confidence, plus permissions and deep links.
- Build comparison
- Diff two builds for new bugs, SDKs, permissions and hosts.
- Screen graph
- Read the map of screens reached during the crawl.
- Optional crawl start
- Starting crawls is off unless you allow it when connecting.
Good for
- Ask an agent to fix the worst bug from the latest QA run
- Audit which analytics hosts load before consent
- Review what changed between two builds before release
- Look up which SDKs a public app ships on Android
Pricing
Paid, from $99 a month
Scout $99/mo, Hunter $499/mo, Enterprise custom; free for early-stage apps
Prices change. Check flytrap.dev before you buy.
FlyTrap FAQ
Can the agent start new crawls?
Only if you tick the option to allow starting crawls when connecting, or enable it when creating an API key.
Do I need an API key?
The hosted server signs in with OAuth. The local package uses FLYTRAP_API_KEY, and the public catalog tools work without one.
Does it test my app itself?
No. The server only reads results that FlyTrap has already produced.
What do plans cost?
Scout is $99 a month and Hunter $499 a month, with custom Enterprise pricing. Apps under 5,000 downloads and under a year old can ask for free access.
Is FlyTrap free?
No. FlyTrap is a paid product. Paid plans start at $99 a month.
How do I add FlyTrap to Claude Code?
Run this command in your terminal: claude mcp add mcp --env FLYTRAP_API_KEY='<FLYTRAP_API_KEY>' -- npx -y @flytrap-qa/mcp
What are the best alternatives to FlyTrap?
The closest alternatives to FlyTrap are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.
Alternatives to FlyTrap

REA
MCP serverCLI and MCP server that lets coding agents reverse engineer apps and binaries
Runs locally- Open source
- 464K GitHub stars
- Developer tools

Codebase Memory
MCP serverLocal MCP server that builds a code knowledge graph for AI coding agents
Runs locally- Open source
- 48.7K GitHub stars
- Knowledge and memory

GitHub
MCP serverOfficial MCP server connecting AI assistants to GitHub repos, issues and workflows
Not checked yet- Remote
- Open source
- 34.5K GitHub stars
- Developer tools

Serena MCP: the IDE for your agent
MCP serverMCP toolkit giving coding agents symbol-level code retrieval and editing
Runs locally- Open source
- 31.5K GitHub stars
- Developer tools

XERJ
MCP serverLocal search engine for AI agents with cited answers over your code and docs
Runs locally- Open source
- 19.2K GitHub stars
- Knowledge and memory

Ouroboros
MCP serverSpec-first layer that pins requirements and verifies results of AI coding agent work
Runs locally- Open source
- 6.4K GitHub stars
- Developer tools