index / RiskAverseTech/toolgate
toolgate
A measured tool-call firewall for AI agents: allow, ask or deny before the agent acts. Claude Code mod and hook, MCP proxy. Built on TypeSafe Jev, measured on real sessions with a blind second labeler.
Agent and Dev ToolingVerificationListedTypeScriptreadme pick
Repository
Stars29
LanguageTypeScript
LicenseMIT
Created2026-09-18
Last pushed2026-10-31
Officialno
Judgment
Genuine Jev project0.88
gate 0.5
Uses Jev at runtime0.84
Is a meta list0.02
Is a reimplementation0.07
Quality scales
Substance2.94 / 3
gate 0.5
Docs2.93 / 3
Novelty2.30 / 3
Composite score0.96
Category distribution
Agent and Dev Tooling0.98
Integrations0.02
SDKs and Clients0.00
Applications0.00
Games and Simulation0.00
Research and Evals0.00
Learning0.00
Other0.00
Assigned categoryAgent and Dev Tooling
Category probability0.98
Category confidence0.98
PatternVerification
Pattern confidence0.99
Curation
StatusListed
Reasongate passed
README pickyes
Overridenone
Sources
- search:jev typesafe
- search:jev in:name,description,readme created:2026-09-18
- list:hellogumbo/awesome-jev
- list:AbdelStark/awesome-typesafe
- list:AnotiaWang/awesome-jev
Provenance
Modeljev-1.14.0
Question setv2
Judged at2026-11-01 12:33 UTC