@xshieldai/chitta-detect
v0.2.1npm· JavaScriptMemory poisoning detection primitives for AI agents — pure pattern matchers (RAG trust, agent-role imperatives, tool-output poisoning, capability expansion, injection fingerprints) + opt-in Agentic Control Center event bus. Extracted from chitta-guard.
The verdict
Maintained. Maintained, actively maintained.
Live from the npm registry · derived rules, not AI
How it scores
MaintenanceHealthy
PopularityUnknown
SecurityClean
LicenseOther
DepsZero deps
Maintenance
Last published this month.
Popularity
Download count unavailable.
Security
No known advisories for this version (OSV).
License
AGPL-3.0-only
Dependencies
No runtime dependencies
Recent releases
- 0.2.1this month
- 0.2.0this month