Appearance
What This Preset Enables
The threat-review preset enables explicit opt-in rules for malicious, credential-bearing, or spyware-like behavior that is useful to detect, but too aggressive to present as part of the quiet default or the softer preview lane.
This includes patterns such as:
- committed hooks or command configs that exfiltrate secrets, dump environments, or execute remote payloads
- committed MCP, plugin-hook, or Claude hook commands that attempt persistence, privilege escalation, or device capture
- instruction surfaces that carry overtly dangerous hidden directives, inline execution payloads, or committed secret-bearing examples
When To Use It
Use it when you want an explicit malicious-behavior review pass on top of the normal product experience, especially for red-team-style audits, suspicious repos, or security triage where spyware-like and post-exploitation patterns should be surfaced intentionally.
Tradeoffs
These rules are strong and useful, but they are intentionally more forceful than the main preview lane. Keeping them explicit helps lintai stay honest with the community about what belongs in the quiet default, what belongs in broader review, and what belongs in a dedicated threat-review pass.