mcpware/IntentProbe

View on GitHub ↗

Activation-probe security scanner for AI agent tooling. Reads a model's internal activations to detect poisoned MCP servers, skills, and packages before install.

9 ★0 forksPythonUpdated 2mo ago

Topics

activation-probingagent-securityai-agent-securityai-securityclaude-codeguardrailsinterpretabilityllm-securitylocal-firstmcpmcp-securitymcp-servermodel-context-protocolopen-sourceprompt-injectionruntime-securityscannersecurity-scannertool-poisoningtool-security
Stars
9★
Forks
0
Language
Python
License
Apache-2.0
Created
2026-06-01
Last push
2026-06-19