OPEN SOURCE · CURATED REPOSITORIES
Curated Jev SDKs, tools, demos, integrations, benchmarks, and research projects. Repository metrics come from GitHub; titles, summaries, and categories are editorial.
Ranked by GitHub Stars
Ranked by recent Star growth and update activity
NyarlathotepppppNyarlathoteppppp/pi-heedRuntime constraints for the pi coding agent: checks every side-effecting tool call against user-stated rules, powered by TypeSafe Jev. Includes a detailed benchmark and experiment log evaluating Jev's System One decisions.
cwjechw98-langcwjechw98-lang/jev-gatesThree guardrail gates for coding agents: pre-action approval, evidence-based completion, and rubric regression checks, powered by TypeSafe Jev (System One). Includes CLI scripts, offline tests, and CI workflows.
jomatsujomatsu/pi-jev-auto-modeA permission gate extension for the Pi coding agent that uses TypeSafe Jev (System One) to semantically approve or block bash, write, and edit tool calls, failing closed by default.
godspedegodspede/construct-auto-classifierEffect-based safety gate for AI coding agents' shell commands, using TypeSafe's Jev as default judge; Jev was the only model certified at zero dangerous commands allowed in this project's evaluation.
lgy1027lgy1027/jevshieldAn AI agent tool-call security gate powered by TypeSafe's Jev model, with LangChain support and zero-config local fallback.
CogFluxCogFlux/opencode-jev-guardAn OpenCode 2 plugin that asks TypeSafe's Jev to evaluate local and FarHand remote shell commands, prompting before risky actions; supports custom risk categories and includes a benchmark.
AskTheWayAskTheWay/dsh-jev-interceptorIntegrates TypeSafe Jev into DeepSeek Harness to classify tool-call risk and auto-approve safe calls in milliseconds, with shadow/enforce modes, telemetry, and 52 tests.
kurihadakurihada/pi-jev-permitA Jev (TypeSafe System One) permission gate for the Pi coding agent. It judges every bash/write/edit call before it runs, using Jev's decision model.
CeamKrierCeamKrier/semantic-firewallA semantic firewall gating LLM tool calls with TypeSafe Jev (System One decision model) plus deterministic policy, including evaluation and stability tests.
sperictaosperictao/dsh-auto-review-jevDeepSeek Harness plugin: per-tool-call Auto-permission review powered by TypeSafe Jev, with account usage and API-key management inline on its settings page.
pantos12pantos12/mailverdictForward an email, get a calibrated phishing verdict. Uses TypeSafe Jev / System One for classification, with MCP server for Copilot Studio and REST for Power Automate.
leepokaileepokai/jev-guardA security hook for coding agents that uses Jev to risk-score tool calls, detect prompt injection, and check skills/plugins. Works with Claude Code, Codex, Copilot, Gemini, Cursor, pi, OpenCode, and ACP.
