questions: { r0_missing: boolean, r0_instruction: boolean, }
yoshi
Context-pruning proxy for Claude Code and Codex: Jev judges which history is still needed, measured not claimed. POC here now, heading soon into https://github.com/compozy/compozy

Every project here really calls Jev in its source · refreshed daily
questions: { r0_missing: boolean, r0_instruction: boolean, }
Context-pruning proxy for Claude Code and Codex: Jev judges which history is still needed, measured not claimed. POC here now, heading soon into https://github.com/compozy/compozy
questions: { actionable: boolean, priority: choice, value: score, }
Open-source Jev log triage for OpenTelemetry. Score the signal before expensive LLM analysis.
questions: { call_t1: boolean, }
Verbatim Jev-scored context reduction for omp, over TypeSafe or OpenRouter
questions: { comparison: choice, }
Jev powered OpenCode compaction
questions: { ping: boolean, }
Codex plugin: Jev-guided dieting of bulky tool results
questions: { q: boolean, }
Jev-powered context compaction for Pi. Keep critical instructions and tool history, prune the noise, and fall back gracefully.
export const SYSTEM_ONE_URL = 'https://api.typesafe.ai/v1/systemone';
Claude Code plugin that replaces the compaction summary with Jev decisions: every tool call and result is scored in one fast request, stale ones are dropped or truncated, everything kept stays verbatim.
from typesafe_sdk import TypeSafeError
The invalidation layer for AI memory. Every fact gets a lease; new evidence ends it. Built on TypeSafe Jev.
DEFAULT_BASE_URL = "https://api.typesafe.ai"
Retrieve by relevance, not resemblance: filter an AI assistant's memories with TypeSafe's Jev
export const SYSTEM_ONE_URL = "https://api.typesafe.ai/v1/systemone";
Pi extension: verbatim context compaction with TypeSafe Jev decisions
export const SYSTEM_ONE_URL = 'https://api.typesafe.ai/v1/systemone'
Jev-scored context compaction for OpenAI Codex CLI — scores every tool call before compaction and restores critical tool outputs verbatim after it
export const SYSTEM_ONE_URL = "https://api.typesafe.ai/v1/systemone";
Fast JEV compaction extension for pi
const ENDPOINT = "https://api.typesafe.ai/v1/systemone";
Reversible context pruning for Pi, powered by TypeSafe Jev. Keep useful context without deleting session history.
const TYPESAFE_URL = "https://api.typesafe.ai/v1/systemone";
Cache-neutral context trimming for the pi coding agent, powered by TypeSafe Jev: long tool output cut to verbatim key lines before it enters context, with lossless recall. Measured, with pre-registered benchmarks.
baseUrl: "https://api.typesafe.ai/v1/systemone",
Verbatim context pruning for the pi coding agent, scored by TypeSafe Jev: stale tool calls and results are dropped or truncated, everything kept stays verbatim.
export const DEFAULT_BASE_URL = "https://api.typesafe.ai";
A Jev based context curator for pi
export const SYSTEM_ONE_URL = "https://api.typesafe.ai/v1/systemone";
Fast compaction for Pi: Jev-scored verbatim eviction of stale tool history, with LLM summarisation of the evicted transcript as the fallback.
// POST https://api.typesafe.ai/v1/systemone
Map a codebase into units and let Jev (TypeSafe AI) hand an AI coding agent the ten files that matter for a task
ENDPOINT = "https://api.typesafe.ai/v1/systemone"
Capability-aware context-pruning installer, bounded Jev assessment engine, and experimental harness request-projection adapters (experimental, not live-host tested)
export const SYSTEM_ONE_URL = 'https://api.typesafe.ai/v1/systemone';
Jev-guided verbatim supplemental compaction for Codex sessions
import type { TypeSafeClient } from '@typesafe-ai/sdk';Context compaction and safety gating for AI agents via TypeSafe Jev: keeps messages verbatim, no summarization. OpenAI, Anthropic, LangChain, CLI, MCP.
export const DEFAULT_MODEL = 'typesafe/jev-1.13';
Claude Code mod: Jev-scored context pruning through OpenRouter. Verbatim compaction, an optional gate on oversized tool outputs, tests and evals.
TYPESAFE_BASE_URL Default https://api.typesafe.ai
TypeSafe Jev-scored context recovery for Cursor CLI (agent). Capture tool I/O, score keep/drop, re-inject after native compact.
baseUrl: "https://api.typesafe.ai/v1/systemone",
Jev-scored context pruning for OpenCode: drops stale tool calls and truncates bulky results on the outgoing request — fail-open, cache-backed, configurable live. Port of fast-jev-compaction to the V2 context hook.
baseUrl: "https://api.typesafe.ai/v1/systemone",
Jev decides what to keep. Compaction never rewrites the transcript.