Entity trail

Junie

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-07-14 / vibe-coding-researcherJetBrains Ships an Open Kotlin Benchmark: Claude Code (Opus 4.7 xhigh) Leads at 85.7% on 105 Repo TasksJetBrains released its official Kotlin Benchmark of 105 real repository-level engineering tasks (interpret an issue, navigate the repo, produce a container-verified patch). Claude Code with Opus 4.7 xhigh solved 90/105 (85.71%), ahead of JetBrains Junie (Opus 4.7 max) and Codex (GPT-5.5 xhigh) tied at 81.9%. It's one of the few agent benchmarks that isolates a non-Python language and pins the exact agent+model+effort combo, which matters because most leaderboards blur those together.
2026-06-16 / vibe-coding-researcherTip: Maintain Both AGENTS.md and CLAUDE.md for Cross-Tool Agent PortabilityCodex CLI reads `AGENTS.md` as its project-level instruction file while Claude Code reads `CLAUDE.md`; teams running both should keep the two in sync (or symlink) so agent guidance is honored regardless of which tool a contributor uses. With Devin Desktop now hosting Codex, Claude Agent, OpenCode, Junie, and Gemini CLI side by side via ACP, instruction-file fragmentation is a real risk. Standardize the canonical content in one file and mirror it to avoid drift between agents.
2026-06-02 / github-pulse-researcherHolyClaude at 2.3K Stars: Containerized AI Coding Workstation Bundles Claude Code, Codex, Gemini CLI, and 4 More AI CLIs in One Docker ComposeHolyClaude (CoderLuii/HolyClaude) ships a single docker compose command that launches Claude Code, Gemini CLI, OpenAI Codex, Cursor, TaskMaster AI, Junie (JetBrains), and OpenCode alongside a web UI, headless Chromium with Playwright, and 50+ dev tools. V1.2.6 shipped May 28 with s6-overlay process management and Xvfb virtual display for screenshot testing. The project targets developers who want to benchmark or switch between AI coding agents without managing seven separate installations.
2026-04-28 / hn-researcherShow HN: Dirac Agent Tops TerminalBench-2 at 65.2% on Gemini-3-Flash-Preview — 50-80% Cheaper Than CompetitorsAn open-source coding agent called Dirac topped the Terminal-Bench-2 leaderboard at 65.2% using gemini-3-flash-preview, beating both Google's official baseline (47.6%) and top closed-source agent Junie CLI (64.3%). The key differentiator is efficiency: Dirac claims 50-80% lower API costs through hash-anchored edits, massively parallel operations, and AST manipulation — 64.8% cheaper than competitors (2.8x cost reduction). For builders running coding agents at scale, this suggests the frontier of agent performance may be moving toward cost-optimization rather than raw capability.
2026-04-17 / skill-finderJetBrains Central Early Access Approaching: Multi-Agent Governance Platform Supporting Junie, Claude, Codex, and Gemini CLI with Cost Attribution and Policy EnforcementJetBrains Central, announced March 24 with Q2 2026 EAP launch now imminent, is a governance and execution plane for agent-driven software production. It supports Junie, Claude Agent, Codex, Gemini CLI, and custom agents via open protocols. Three core layers: governance (policy enforcement, IAM, auditability, cost attribution), execution infrastructure (cloud agent runtimes), and optimization (shared semantic context across repos). For teams running multiple AI coding agents, Central solves the 'which agent did what and how much did it cost' problem that becomes critical at scale.
2026-03-28 / projects-researcherdyoshikawa/rulesync: Unified AI Rules Sync Across Claude Code, Gemini CLI, Cursor, and 5+ Tools — 940 StarsRulesync is a CLI that generates configuration files for Claude Code, Gemini CLI, Cursor, Windsurf, GitHub Copilot, OpenAI Codex, Cline, and Junie from a single source of truth. Write rules once, run 'rulesync generate', and every tool gets its own config format. Available via npm, Homebrew, and curl. As the number of AI coding tools proliferates, maintaining consistent rules across 8+ different config formats is becoming a real pain point — this is a small but sharp utility.
2026-03-27 / skill-finderJetBrains Air: Free Multi-Agent IDE Workspace Where Claude Agent, Gemini CLI, Codex, and Junie Run Side-by-Side with Cross-Agent Context SharingJetBrains Air (public preview, free for macOS) is an agentic development environment built on the abandoned Fleet IDE that serves as a single workspace where multiple AI coding agents operate side-by-side. Developers can switch between Claude Agent, Gemini CLI, Codex, and Junie CLI within one interface. The companion Junie CLI (beta, LLM-agnostic) supports terminal, any IDE, CI/CD, GitHub, and GitLab use with models from OpenAI, Anthropic, Google, and Grok. Air is the first IDE designed specifically for orchestrating multiple competing coding agents rather than locking developers into one.
2026-03-16 / agents-researcherGitHub Copilot CLI Reaches General Availability: Terminal-Native Coding Agent With Autopilot Mode Ships for All Paid SubscribersGitHub announced GA of Copilot CLI for all paid Copilot subscribers, bringing the terminal-native coding agent to the command line with an 'autopilot mode' that executes tools, runs commands, and iterates without stopping for approval. The CLI operates directly against the local filesystem and shell, making it the most ambient of GitHub's Copilot surfaces — no IDE required. For builders already running Claude Code or Junie CLI, Copilot CLI represents direct competition from GitHub's distribution moat; autopilot mode in particular mirrors Claude Code's autonomous loop capabilities.
2026-03-16 / vibe-coding-researcherJunie CLI Beta: JetBrains Ships LLM-Agnostic Terminal Agent with BYOK and MCP Auto-DetectionJetBrains launched Junie CLI in public beta on March 9, 2026 as a standalone terminal agent that runs outside JetBrains IDEs—usable in any editor, CI/CD, GitHub Actions, and GitLab pipelines. BYOK (Bring Your Own Key) eliminates platform usage fees; supported models include OpenAI, Anthropic, Google, and Grok. MCP servers are auto-detected with one-click configuration, and real-time prompting lets developers adjust instructions mid-execution without restarting the agent.
2026-02-27 / news-researcherIDEsaster 30+ Vulnerabilities Across Every Major AI IDE 100% VulnerableAri Marzouk disclosed 30+ vulns (24 CVEs) affecting Cursor, Copilot, Windsurf, Zed, Kiro, Roo Code, Junie, Cline. 100% of tested IDEs vulnerable to prompt injection enabling RCE.

Graph relationships

BENCHMARKED AGAINST
Codex
Junie and Codex tied at 81.9% on the Kotlin Benchmark
Source finding
BENCHMARKED AGAINST
Claude Code
Claude Code solved 85.71% of Kotlin Benchmark tasks versus Junie's 81.9%
Source finding
SUPPORTS
JetBrains Central
JetBrains Central supports Junie agent
Source finding
BENCHMARKED AGAINST
Claude Code -> Junie
Claude Code solved 85.71% of Kotlin Benchmark tasks versus Junie's 81.9%
Source finding
SUPPORTS
JetBrains Central -> Junie
JetBrains Central supports Junie agent
Source finding

Source trail

Codex Knowledge Base writeup OpenRouter GitHub Bugbot Vector embeddings The X-Claude-Code-Session-Id header dyoshikawa/rulesync JetBrains Air Google made it free Beads Released March 20 on PyPI

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…