ft-scout@4.0.8
FrontTerrain — The Onboarding Co-Pilot for New Codebases
AI Security Review
scanned 3mo ago · by lpm-firewall-aiNo concrete malicious install-time behavior, hidden payload execution, or unconsented credential exfiltration was found. The package nevertheless exposes a high-impact remote-agent capability with weak execution boundaries.
Static reason
High-risk behavior combination matched malicious policy.; previous stored version diff introduced dangerous source
Trigger
User runs `scout agent` or `scout setup`.
Impact
A compromised or unsafe remote model response could modify files or execute commands in the user’s workspace.
Mechanism
User-invoked remote AI agent with filesystem and shell capabilities.
AI rationale
No concrete malicious install-time behavior, hidden payload execution, or unconsented credential exfiltration was found. The package nevertheless exposes a high-impact remote-agent capability with weak execution boundaries.
Evidence
package.jsonbin/ft.jsbin/src/index.jsbin/src/engine/agentEngine.jsbin/src/engine/llm.jsbin/src/commands/setup.jsbin/src/utils/email.jsbin/src/utils/firebaseAuth.js
Network endpoints1
frontterrain.com/api
Decision evidence
public snapshotAI called this Suspicious at 90.0% confidence as Dangerous Capability with low false-positive risk.
Evidence for warning
- `scout agent` sends workspace context to FrontTerrain’s LLM proxy and exposes file-write and shell-command tools.
- `scout setup` explicitly runs dependency-install commands in the selected project.
- `test-email` can use local SMTP/API environment credentials to send mail when explicitly invoked.
Evidence against
- `package.json` has no preinstall, install, or postinstall hook.
- Networked login/signup submits user-entered credentials to the declared Firebase tenant.
- Email dispatch is reached only by explicit signup or `test-email`; no secret-harvesting or credential-exfiltration path was found.
- Auto-upgrade requires an interactive confirmation before global npm installation.
Behavioral surface
ChildProcessEnvironmentVarsFilesystemNetworkShell
HighEntropyStringsUrlStrings
Source & flagged code
13 flaggedbin/src/utils/firebaseAuth.jsView file
1patternName = google_api_key
severity = high
line = 1
matchedText = export c...lg';
High
High Secret
Package contains a high-severity secret pattern.
bin/src/utils/firebaseAuth.jsView on unpkg · L11patternName = google_api_key
severity = high
line = 1
matchedText = export c...lg';
High
Secret Pattern
Google API key in bin/src/utils/firebaseAuth.js
bin/src/utils/firebaseAuth.jsView on unpkg · L1bin/src/utils/updateCheck.jsView file
5import { confirm, spinner, note } from '@clack/prompts';
L6: import { execSync } from 'child_process';
L7: const CACHE_FILE = path.join(os.homedir(), '.scout-update-cache.json');
High
Child Process
Package source references child process execution.
Findings
3 Critical8 High4 Medium5 Low
CriticalCredential Exfiltrationbin/src/utils/email.js
CriticalTrigger Reachable Dangerous Capabilitybin/src/utils/email.js
CriticalPrevious Version Dangerous Deltabin/src/engine/verifier.js
HighHigh Secretbin/src/utils/firebaseAuth.js
HighChild Processbin/src/utils/updateCheck.js
HighShellbin/src/commands/agent.js
HighSandbox Evasion Gated Capabilitybin/src/commands/setup.js
HighTrigger Reachable Credential Exfiltrationbin/src/utils/email.js
HighRuntime Package Installbin/src/utils/updateCheck.js
HighShips High Entropy Blobfront-terrain-cli-1.0.0.tgz
HighSecret Patternbin/src/utils/firebaseAuth.js
Affected versions and remediation
This report applies to ft-scout@4.0.8.
See all versions of ft-scout for other recorded verdicts.
- Review the evidence and your use of ft-scout@4.0.8 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.