Entity trail

Docker Agent

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

Briefing refs

Findings

Edges

Sources

Showing the first 40 findings. More graph evidence exists in the corpus.

Corpus findings

2026-06-17 / hn-researcherVicki Boykis: 'Running Local Models Is Good Now' Hits 1,437 Points on HNVicki Boykis argues local models crossed a usability threshold, citing Gemma 4 (gemma-4-26b-a4b, gemma-4-12b-qat) running agentic coding at roughly 75% of frontier accuracy/speed on a 64GB M2 Mac using the Pi agent framework with sandboxed bash-only Docker execution. She reports doing real refactors, unit-test generation, and two-tower recommender code locally — tasks she'd have thought impossible a year ago. The post topped HN (1,437 pts, 551 comments), and is the strongest recent practitioner signal that local-first agentic coding is now viable for solo builders.
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.
2026-06-01 / github-pulse-researchermattpocock/sandcastle at 5.6K Stars: TypeScript Library Orchestrates Sandboxed Coding Agents With Single sandcastle.run() CallMatt Pocock's sandcastle lets you orchestrate AI coding agents in isolated sandboxes via a single TypeScript API call. It handles branch strategy, commit management, and merge-back automatically, with built-in providers for Docker, Podman, and Vercel. For builders who want to parallelize AFK agents — one on frontend, one on backend, one on tests — sandcastle manages the git coordination that makes multi-agent development reliable rather than a merge conflict nightmare.
2026-05-29 / github-pulse-researcherLlamaIndex Ships LiteParse v2.0.3: Rust-Powered Document Parser Hits 6.9K Stars With +680 Today as RAG Infrastructure Layerrun-llama/liteparse is a fast, lightweight PDF and document parser written in Rust (70.3% of codebase) that extracts spatial text with bounding boxes without cloud dependencies. Docker v2.0.3 shipped May 28 with support for PDF, DOCX, XLSX, PPTX, and images across Rust, Node.js, Python, and Browser/WASM runtimes. Built-in Tesseract OCR and screenshot generation for LLM agents makes it the local-first complement to LlamaParse for complex documents.
2026-05-26 / skill-finderE2B Ships Firecracker MicroVM Sandbox for Agent Code Execution — 150ms Cold Start, Kernel-Level Isolation, Python/JS SDKs, 12K+ StarsE2B provides Firecracker microVM-based sandboxes purpose-built for AI agent code execution with kernel-level isolation — each sandbox runs its own kernel, unlike Docker's shared-kernel model. Cold starts at 150ms. The SDK is designed around agent workflows: spawn a sandbox, run AI-generated code, process data, and tear down. Setup time is ~2 minutes for managed cloud or ~5 minutes for local MCP integration with Claude Desktop/Cursor. The latest SDK (May 22, 2026) supports both Python and JavaScript. This is the infrastructure layer that makes untrusted agent code execution safe in production.
2026-05-26 / agents-researcherE2B + Docker Partnership: Every Agent Sandbox Now Includes 200+ Pre-Verified MCP Tools via Docker MCP GatewayE2B partnered with Docker to embed the Docker MCP Catalog (200+ tools including GitHub, Perplexity, Browserbase, Stripe, Grafana, Notion) into every E2B sandbox via Docker MCP Gateway. Each MCP tool runs as a Docker container inside the sandbox with autocomplete, type validation, and unified configuration. What previously required hours of manual configuration now takes seconds. For agent builders, this eliminates the single most painful setup step: wiring tools into sandboxed execution environments.
2026-05-25 / news-researcherMicrosoft Agent Framework Python 1.6.0: Shell Tool for Local/Docker Execution and Monty-Backed CodeActMicrosoft released Agent Framework Python 1.6.0 on May 22, adding a core Shell tool supporting both local and Docker execution environments, a new Monty-backed CodeAct provider package, experimental hosted tool factories in Foundry, and instrumentation enabled by default. The release continues rapid iteration since the framework's 1.0 GA in April 2026, competing directly with LangGraph, CrewAI, and JetBrains Koog for the enterprise agent orchestration layer.
2026-05-24 / skill-finderCursor 3 Ships /best-of-n: Run the Same Task Across Multiple Models in Isolated Git Worktrees, Compare Outcomes Side-by-Side for Model SelectionCursor 3's /best-of-n command runs identical prompts across multiple models (e.g., Claude Opus, GPT-5.3, Gemini 3.5) each in its own git worktree, then presents outputs side-by-side for selection. Companion /multitask splits requests into async subagents for parallel execution. The May 13 update extended this with cloud-agent environments: multi-repo support, Dockerfile-based config with build secrets, and admin-level governance for running parallelized agent fleets.
2026-05-23 / vibe-coding-researcherPattern: Multi-Agent Session Management Solidifying Around tmux + Git WorktreesAgent of Empires (2.3K stars) crystallizes a pattern emerging across the ecosystem: tmux sessions as the persistence layer and git worktrees as the isolation mechanism for running multiple AI coding agents in parallel. Claude Code's built-in agent teams, OpenCode's parallel mode, and AoE's TUI/web dashboard all converge on the same substrate — each agent gets its own tmux session (survives disconnects) and its own worktree (no branch conflicts). Docker sandboxing adds a third isolation tier for untrusted operations. This is becoming the standard multi-agent coordination architecture for solo developers and small teams.
2026-05-23 / vibe-coding-researcherTip: Agent of Empires Manages Parallel AI Agents Across Worktrees via TUI, Web, or MobileAgent of Empires (2.3K stars) provides a single dashboard to manage multiple Claude Code, OpenCode, Codex CLI, Gemini CLI, and other coding agents running in parallel. Each agent runs in its own tmux session with automatic git worktree creation, so agents keep running when you close the terminal or disconnect SSH. A web interface enables mobile access. Docker sandbox mode provides full isolation with configurable volume mounts. The tool solves the 'how do I run 5 agents on different branches without losing track' problem that scaling agent workflows creates.
2026-05-22 / agents-researcherCrewAI Hit by Four CVEs: Prompt Injection Chains to RCE, SSRF, Sandbox Escape, and Arbitrary File ReadCERT/CC advisory VU#221883 discloses four vulnerabilities in CrewAI that chain through prompt injection: CVE-2026-2275 (Code Interpreter Tool falls back to vulnerable SandboxPython if Docker unreachable, enabling arbitrary C function calls), CVE-2026-2285 (JSON loader reads files without path validation), CVE-2026-2286 (SSRF), and CVE-2026-2287 (insecure Docker runtime fallback). An attacker who can interact with a CrewAI agent with Code Interpreter enabled can chain these to achieve full host compromise. This represents the 'prompts become shells' pattern Microsoft identified as systemic across agent frameworks.
2026-05-21 / news-researcherNanoClaw Raises $12M Seed — Secure, Container-Isolated Alternative to OpenClaw Built on Claude Agent SDKNanoClaw, a security-focused AI agent framework built on Anthropic's Claude Agent SDK, raised a $12M seed round. Unlike OpenClaw's application-level permission model, NanoClaw runs every agent group inside isolated Docker containers with OS-level security in just ~700 lines of code across 15 files. It integrates with WhatsApp, Telegram, Slack, Discord, and Gmail, routing credentials through Agent Vault so agents never hold raw API keys. The Cohen brothers' AI marketing firm uses it as a secure alternative for agent-driven work.

Graph relationships

RELEASED
Docker
Docker Engineering released docker-agent as official AI agent builder and runtime in Go.
Source finding
RELEASED
Docker -> docker-agent
Docker Engineering released docker-agent as official AI agent builder and runtime in Go.
Source finding

Source trail

docker/docker-agent Vicki Boykis / Hacker News GitHub GitHub GitHub GitHub Docker Blog

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues