882 repositories with documented relationships and source evidence.
Automatic tier-based model routing for DeepSeek Harness (dsh): a virtual `smart` model classifies every request by difficulty (hard / normal / easy) and by vision need, then delegates it to the models you already configured.
Autonomous Jev pull-request review with typed decisions, calibrated approval gates, and trusted-owner escalation
Jev-powered model-judged permission gate for OMP (TypeSafe System One)
Jev, OpenAI Agent, MCP를 활용해 구매·구독·비용 의사결정을 계산·비교·검증하는 프로젝트
Typed judgment layer for coding agents — gates from PRD to ship. Jev-ready, provider-agnostic.
The agent that reads its own mistakes — a self-diagnosing, self-improving ML loop (Weave · W&B Inference · TypeSafe · marimo · ARIA). CoreWeave Hacks 2026.
Send Pi agents back to work when they stop before the job is done.
HSM runtime and AXI for deterministic AI workflows
PoC: Claude agent + Firecrawl that maps misspelled Swiss company names to their UID (CHE number), with confidence score and rationale.
Jev4JobHunter — AI job application plugin with TypeSafe Jev for job matching. Screen BOSS Zhipin jobs, apply and follow up with Codex / Claude Code. 用 Jev 筛岗位，让 Agent 完成投递与跟进。
A Pi coding-agent extension that routes each prompt to the best backend model — cheap, fast, balanced, or powerful — using local heuristics, with optional TypeSafe Jev classification.
agent patterns lab
Verify before you trust. Cross-provider verification with LLMs and Jev. CLI, library & MCP.
Multi-agent LLM investment research assistant (paper/backtest only)
ai design system
Curated Pi coding agent configuration with Jev compaction, LSP, subagents, plan mode, safety tools, and automatic updates.
Collection of custom skills for Claude Code.