jev plugins

by socai-io

Run read-only Instagram, TikTok, and LinkedIn research through Jev-directed socai CLI operations in the user's local browser.

by shitianfang

Hands the agent steps that need no text output to Jev, TypeSafe's judgment model: typed yes/no, pick-one and score questions about one state answered in a single batched call, with typed escalation back to the LLM for anything it cannot or should not decide.

by nahid

Routes each request to one of 27 specialist agent roles, which load only the skills the work needs and verify what they claim.

by Alberto Leal

Find and remove redundant comments and docstring content while retaining rationale, contracts, and documentation/tooling functions. Apply a source-based keep/delete/reduce/rewrite review; optionally consult Jev on redundancy and whether a proposed edit loses useful information. Reuses the configurable general Jev helper and skips it without helper or credentials. Use when cleaning up AI-slop comments, auditing a diff, reducing mixed docstrings, or checking a comment rewrite.

by Alberto Leal

Review and improve engineering Markdown designs, analyses, implementation plans, and ADRs. Measure sections and protected content with a Markdown parser; optionally use Jev for typed quality and meaning-preservation judgments. Preserve requirements, uncertainty, code, and links while the agent makes targeted edits. Jev is optional and provider configurable.

by Alberto Leal

Consult Jev for bounded typed second opinions during agent work: compare plans, check claims against supplied evidence, prioritize investigations, and challenge conclusions. Uses Vercel AI Gateway by default with configurable providers; skips when the selected key is absent. Includes a standalone uv-run helper and offline contract tests.

by Alberto Leal

Measure Python complexity per function and per call path with ruff and complexipy; optionally add Jev judgments through configurable providers (Vercel by default). Use when prioritizing refactors, investigating over-layering, or verifying that a refactor improved the whole path. Without the selected provider key, semantic review is skipped and static analysis continues. Tests verify behavior.

by Ghaleb Dweikat

A calibrated context sieve for Claude Code. Every large tool result is judged by a System One model before it enters context; irrelevant blocks are replaced with a summary and a recall key.

by 0x7067

Hands the small judgments in a session to TypeSafe's Jev, a System One model that returns typed answers instead of generating text: what kind of request each prompt is, which model tier a subagent needs, whether an edit breaks a rule already written in your CLAUDE.md or AGENTS.md, and which transcript blocks survive compaction. Kept blocks carry over verbatim — no LLM writes a summary.

by 0x7067

Jev drives a real browser: one goal, an indexed action space, one TypeSafe systemOne decision per step. Exposes the jev_browse MCP tool.

by Lm09

Semantic code search and call-graph analysis inside Claude Code, powered by the sgrep CLI (Jev / TypeSafe System One). Finds code by meaning, traces flows (callers + callees), and computes blast radius. Requires the sgrep CLI installed separately.

by x96x64

Keeps what matters through Claude Code's own compaction — scores your session's context with Jev right before compaction runs, then hands the highest-scoring entries back as a reminder the moment it finishes. No configuration required; set an explicit goal with /ctxjev:set-goal when you want scoring aimed at something specific.

by Li-ship-dot

Portable Agent MCP plugin for on-demand, verbatim Jev-guided transcript compaction.

by Falconiere Barbosa

Typed judgments from TypeSafe's Jev model at runtime (skill + REST wrapper) — yes/no probabilities, choices, and scores a script can branch on. Standalone, no dependencies.

by Harshit Badiger

Maps where a codebase makes, avoids, or could make semantic decisions, finds the ones worth a System One model, and rejects Jev where code, an LLM, embeddings, classical ML, or a human fits better. Then validates and implements the survivors.

by Oleg Alshaev

A second opinion on every command, edit and MCP call: Jev typed probabilities, a zero-key allowlist, a local audit log, and a prompt-injection sentinel for Claude Code

by Jev-decision

Build AI-powered software with TypeSafe System One and Jev using typed judgments, calibrated probabilities, and composable decision primitives.

by Jev-decision

Build AI-powered software with TypeSafe System One and Jev using typed judgments, calibrated probabilities, and composable decision primitives.

by jev-enhanced-plugin contributors

Third-party skills enhanced with advisory TypeSafe Jev judgments: two-axis code review, prose humanizing (en/zh-TW), reply checking, simplification gating, and design-tree grilling, plus a procedure for migrating further skills.

by Kadihx

Offline $0 decision layer for coding agents: Choice/Score/Noul primitives, a BELKI confidence gatekeeper, ultra-planning, red-teaming, 4-channel research, lossless log compaction and RLVR self-improvement — as an MCP server + CLI + skill.

by Alexei Ledenev

Auto-picks the right Claude model for each turn — small models for quick edits, mid-tier for code, frontier for hard problems. Uses Jev to classify each request.

by Falconiere Barbosa

Host-native project status: a persistent Claude Code statusline plus an explicit Codex status skill for repository, gate, git, comemory, and Jev readiness. Standalone, no dependencies.