Agent Tooling for Jev

TradingAgents combines fundamental, sentiment, news and technical analysts with debating researchers, a trader and risk management roles. With TYPESAFE_API_KEY, Jev removes unrelated StockTwits and Reddit posts and classifies stance; without the key, posts pass through unscreened.

AgentScope builds agents around model reasoning and tool use, with context management, event streaming, memory and permission controls. JevClassifierModel integrates the Jev classifier through the TypeSafe SDK within this broader agent framework.

Planning and execution agents gather, summarize and track sources, then combine findings into a cited research report. When TYPESAFE_API_KEY is present, Jev scores passage usefulness for the research question before the LLM reads the context.

This self-hosted trading stack covers Python strategies, backtests, paper trading, live execution and monitoring. Jev evaluates order, strategy, exposure, position and budget context, returning typed choices with probabilities for the entry filter.

The self-hosted platform organizes agents, isolated workspaces and task boards across your own computers. The README lists TypeSafe's Jev as an available model that returns typed decisions with confidence rather than generated text.

This npm library and Claude Code plugin compact history by pruning tool calls and results instead of rewriting conversation text. Jev answers separate noul questions about retaining each call and result; thresholds determine whether to keep, shorten or remove the pair.

The native CLI supports an interactive shell, scripted requests and embedding, with skills, MCP servers and subagents. Jev receives review policy, context and the pending action, returning a clear or caution decision for the permission pipeline.

Searchable memory helps recover past decisions, debugging context and feature history across supported coding agents; Markdown remains the source of truth. Optional Jev reranking uses the remote TypeSafe API to reorder memory search results without downloading a local reranker model.

Teams collaborate with multiple agents across chat and development platforms, starting work from messages, issues, pull requests, webhooks or schedules. Jev helps choose the agent and model, including routing support conversations to the appropriate specialist.

The jg CLI helps coding agents explore unfamiliar repositories by returning relevant files, line references and source excerpts. Jev judges folders, file previews and declarations as the search follows relevant branches through the repository.

Reticle checks running apps using network, console and state evidence, returning pass, fail or unknown. The optional Jev exploration driver selects DOM candidates; internal verification-flow routing remains planned. Connect a development app through MCP, describe expected behavior and save flows for deterministic replay.

Hermes Jev Skills provides reusable skills for model routing, search, memory filtering, triage and browser actions. Jev makes bounded choices such as selecting a model or ranking retrieved passages; the host model handles prose. The skills work with Hermes, Claude Code and Codex. Shadow mode records routing decisions without switching models.

hippo-memory stores agent memories in SQLite, tracks provenance and lets feedback suppress incorrect memories. Its optional Jev reranker orders recalled memories and is disabled by default. Agents can remember and recall through the CLI or MCP, sharing memory across coding tools.

DSCODE is a macOS coding agent whose terminal interface, CLI and scripts share a persistent session runtime. The repository includes a separate Jev client for OpenRouter’s alpha Decisions endpoint, supporting Choice, Score and Noul. Use sessions for editing, tests and agent handoffs; the Jev client requires an API key.

Distill is a terminal coding harness supporting Grok, Codex subscriptions and OpenAI-compatible models. Its Jev controls cover model tiers, skill suggestions and memory ranking; permissions remain the harness’s responsibility. Configure main, reasoning and utility models in Model tiers; the main model runs every session and step.

Jev Review scans Git diffs or complete codebases for correctness, security, reliability, compatibility and test coverage. Jev selects evidence, classifies mechanisms, scores severity and, when needed, uses Choice to select a reviewer. Code applies thresholds and workflow policy. Use Node.js 24+, Git and a TypeSafe key to generate JSON reports or inspect saved reviews locally.

Foreman supervises coding workers, assessing completion, requirements, test sufficiency and the need for human input. Jev evaluates checks; a deterministic Python arbiter selects whether to continue, steer, verify, stop or escalate. Supply a repository and job with Python 3.11+, a TypeSafe key and a configured coding-agent backend.

The collection supplies skills for record triage, evidence checks, rubric evaluation and choosing legal actions. Follow the installation guide, choosing real Jev through TypeSafe or OpenRouter, or an explicitly labeled agent simulation.

phi is a Go terminal coding harness with file tools, subagents, diff review and on-demand MCP tool discovery. Configure a primary coding model with phi config, start phi, and supply TYPESAFE_API_KEY if enabling optimizer features.

jev-router launches the native coding CLIs through a local proxy and selects a model tier for each new user turn. Install jev-router with npm, configure JEV_API_KEY, and launch jev-claude or jev-codex with an installed, authenticated upstream CLI.

VexJoy Agent connects requests to specialist agents, reusable skills and workflows across supported coding runtimes. Install the toolkit for the desired runtime and configure direct TypeSafe or Vercel transport; the installer supports a dry-run preview.

Mobile Jev controls an Android device through Mobilerun, with a live studio, CLI and execution traces. Configure both provider keys and a ready device ID, then use pnpm dev to open the local studio and submit a goal.

JevHarness lets a language model write task-specific features, questions and control logic, with optional improvement from rewards and complete traces. Explore the archived Pokémon run locally with the website scripts, or install the skill and describe a task to build a harness.

LLM Debugger drives VS Code’s debugger, setting breakpoints, stepping and inspecting live variable values. The Jev strategy selects next steps and checks evidence sufficiency; a generation model supplies hypotheses, expressions and fixes. Configure model keys and use @debugger with a program or symptom; candidate programs are run to reproduce it.