{
  "listings": [
    {
      "slug": "claude-code",
      "name": "Claude Code",
      "company": "Anthropic, PBC",
      "grade": "C",
      "fit": "yes",
      "why": "claude --worktree \u003cname\u003e starts each session in its own git worktree under .claude/worktrees, and claude agents --json lists background sessions for a script.",
      "source": "https://code.claude.com/docs/en/cli-reference",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/claude-code"
    },
    {
      "slug": "cursor-cli",
      "name": "Cursor CLI",
      "company": "Anysphere, Inc.",
      "grade": "E",
      "fit": "yes",
      "why": "-w, --worktree [name] runs each task in a new worktree under ~/.cursor/worktrees, --worktree-base picks the base, and -p runs exit on their own.",
      "source": "https://cursor.com/docs/cli/reference/parameters",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/cursor-cli"
    },
    {
      "slug": "devin",
      "name": "Devin",
      "company": "Cognition AI, Inc.",
      "grade": "C",
      "fit": "yes",
      "why": "devin_session_create starts one or more sessions, each in its own VM, and devin_session_gather tells which settled, errored or are waiting.",
      "source": "https://docs.devin.ai/work-with-devin/devin-mcp",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/devin"
    },
    {
      "slug": "droid-cli",
      "name": "Droid CLI",
      "company": "The San Francisco AI Factory, Inc.",
      "grade": "C",
      "fit": "yes",
      "why": "droid exec --worktree \u003cname\u003e runs each job in its own worktree, the docs show two jobs fanned out in parallel, and clean worktrees are removed on exit while dirty ones stay for review.",
      "source": "https://docs.factory.com/droid-exec/overview",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/droid-cli"
    },
    {
      "slug": "emdash",
      "name": "Emdash",
      "company": "General Action, Inc.",
      "grade": "C",
      "fit": "yes",
      "why": "Add Task creates one or more worktrees, each running its own agent from 34 supported CLIs, with diff review, pull requests and CI checks in the app, locally or over SSH.",
      "source": "https://emdash.sh/docs",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/emdash"
    },
    {
      "slug": "paperclip",
      "name": "Paperclip",
      "company": "Paperclip Labs, Inc.",
      "grade": "C",
      "fit": "yes",
      "why": "Isolated execution workspaces give each task a new git worktree on its own branch, several agents work at once, and task status shows which finished.",
      "source": "https://docs.paperclip.ing/guides/projects-workflow/workspaces/",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/paperclip"
    },
    {
      "slug": "paseo",
      "name": "Paseo",
      "company": "Mohamed Boudra",
      "grade": "B",
      "fit": "yes",
      "why": "`paseo run --new-workspace worktree --worktree-mode branch-off --new-branch \u003cname\u003e` gives each agent its own worktree, `paseo ls` lists running agents, and agents can launch other agents across providers.",
      "source": "https://paseo.sh/docs/cli",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/paseo"
    },
    {
      "slug": "sculptor",
      "name": "Sculptor",
      "company": "Imbue, Inc.",
      "grade": "C",
      "fit": "yes",
      "why": "Each workspace is a git worktree on its own branch, several agents and workspaces run at once, and the app shows changes and a pull request button per workspace.",
      "source": "https://github.com/imbue-ai/sculptor/blob/main/docs/help/workspaces.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/sculptor"
    },
    {
      "slug": "t3-code",
      "name": "T3 Code",
      "company": "T3 Tools Inc.",
      "grade": "C",
      "fit": "yes",
      "why": "Each thread can work in its own worktree, outside agents launch and wait on threads over MCP, and Git actions commit, push and open a pull request.",
      "source": "https://github.com/pingdotgg/t3code/blob/main/README.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/t3-code"
    },
    {
      "slug": "traycer",
      "name": "Traycer",
      "company": "Traycer AI",
      "grade": "C",
      "fit": "yes",
      "why": "`traycer worktree create --workspace --branch` and the app give each agent its own worktree and branch, agents message each other across tasks, and notifications say which agent finished or failed.",
      "source": "https://docs.traycer.ai/cli/commands.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/traycer"
    },
    {
      "slug": "amp",
      "name": "Amp",
      "company": "Amp Frontier Corporation",
      "grade": "C",
      "fit": "caveat",
      "why": "amp -ox or --executor runner:\u003cid\u003e starts a thread elsewhere and prints its URL, so tasks can fan out to orbs or runners, but there is no worktree flag for parallel local runs.",
      "source": "https://ampcode.com/docs/markdown/cli/execute-mode",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/amp"
    },
    {
      "slug": "clawboo",
      "name": "Clawboo",
      "company": "Sanreds",
      "grade": "B",
      "fit": "caveat",
      "why": "Each task-runner task gets its own git worktree, the Board shows running, blocked and Needs you, and a task can pass between Claude Code, Codex, Hermes and OpenClaw, but every child runs with permissions bypassed and no sandbox.",
      "source": "https://docs.claw.boo/llms-full.txt",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/clawboo"
    },
    {
      "slug": "cli-manager",
      "name": "CLI Manager",
      "company": "solhun (Jeong Kyeonghun)",
      "grade": "BB",
      "fit": "caveat",
      "why": "One-click worktree workspaces each with their own sessions, and the API reports idle, awaiting_input and exited per session, but worktree creation is UI only and Git functions are not exposed through the API.",
      "source": "https://github.com/woorichicken/CLI_manager",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/cli-manager"
    },
    {
      "slug": "cline",
      "name": "Cline",
      "company": "Cline Bot Inc.",
      "grade": "C",
      "fit": "caveat",
      "why": "--worktree creates a detached worktree per task and -z runs sessions in the background hub, but every child runs with auto-approval on and there is no sandbox.",
      "source": "https://github.com/cline/cline/blob/main/docs/cli/cli-reference.mdx",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/cline"
    },
    {
      "slug": "conductor",
      "name": "Conductor",
      "company": "Conductor",
      "grade": "E",
      "fit": "caveat",
      "why": "Isolated workspaces on git worktrees run several agents in parallel with review and merge, but nobody new can download it and existing users lose it in early 2027.",
      "source": "https://www.conductor.build/docs",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/conductor"
    },
    {
      "slug": "deepseek-harness",
      "name": "DeepSeek Harness (dsh)",
      "company": "DeepSeek (the LICENSE names DeepSeek, and no legal entity is named in the repository or the docs)",
      "grade": "D",
      "fit": "caveat",
      "why": "The headless README documents running from a worktree with --add-dir granting the shared .git directory, but dsh creates no worktrees itself.",
      "source": "https://github.com/deepseek-ai/deepseek-harness/blob/master/packages/bundle/headless/README.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/deepseek-harness"
    },
    {
      "slug": "deepseek-superagent",
      "name": "dsh-superagent (DeepSeek)",
      "company": "DeepSeek (the LICENSE names the DeepSeek Harness Team; DeepSeek Harness's terms name Hangzhou DeepSeek Artificial Intelligence Co., Ltd.)",
      "grade": "E",
      "fit": "caveat",
      "why": "Parallel Executor Sessions each in a worktree under a concurrency limit, each ending in its own pull request, but every Executor runs without approval as the operator's user.",
      "source": "https://github.com/deepseek-ai/dsh-superagent-prototype/blob/main/docs/003-execution.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/deepseek-superagent"
    },
    {
      "slug": "gemini-cli",
      "name": "Gemini CLI",
      "company": "Google LLC",
      "grade": "B",
      "fit": "caveat",
      "why": "Git worktrees give each session its own copy of the codebase, but the feature is experimental and must be enabled in settings.",
      "source": "https://geminicli.com/docs/cli/git-worktrees/",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/gemini-cli"
    },
    {
      "slug": "github-copilot-cli",
      "name": "GitHub Copilot CLI",
      "company": "GitHub, Inc.",
      "grade": "C",
      "fit": "caveat",
      "why": "--worktree runs or resumes a session in a worktree and --fleet fans out subagents, but exit codes are undocumented so an orchestrator reads the JSON result record instead.",
      "source": "https://docs.github.com/en/copilot/reference/copilot-cli-reference/cli-command-reference",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/github-copilot-cli"
    },
    {
      "slug": "harness-autonomous",
      "name": "Harness",
      "company": "Autonomous, Inc.",
      "grade": "B",
      "fit": "caveat",
      "why": "A worktree per harness, harness new for Claude Code or Codex and harness wait telling idle, needs the person or stopped, but it needs a browser sign-in and an experimental setting, and Codex runs Harness commands outside its sandbox.",
      "source": "https://github.com/autonomous-ai/openharness",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/harness-autonomous"
    },
    {
      "slug": "harnessrouter",
      "name": "HarnessRouter",
      "company": "Lumentree Corporation",
      "grade": "D",
      "fit": "caveat",
      "why": "Codex, Claude Code, Hermes and Pi run behind one contract and a task moves between them by changing the ID, each in its own sandbox, but there are no git branches or diffs.",
      "source": "https://harnessrouter.ai/llms-full.txt",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/harnessrouter"
    },
    {
      "slug": "kilo-code-cli",
      "name": "Kilo Code CLI",
      "company": "Kilo Code Inc. (the terms are a contract with Anaconda, Inc. on behalf of itself and its affiliates, including Kilo Code Inc.)",
      "grade": "BB",
      "fit": "caveat",
      "why": "--worktree creates or reuses a named worktree and the agent_manager tool starts worktree sessions, but --auto approves every request not denied and the sandbox is off by default.",
      "source": "https://github.com/Kilo-Org/kilocode/blob/main/packages/kilo-docs/pages/code-with-ai/platforms/cli-reference.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/kilo-code-cli"
    },
    {
      "slug": "mistral-vibe-cli",
      "name": "Mistral Vibe CLI",
      "company": "Mistral AI (RCS Paris 952 418 325)",
      "grade": "B",
      "fit": "caveat",
      "why": "`--worktree \u003cname\u003e` gives each run its own checkout under `~/.vibe/worktrees/` and exit codes report the end, but there is no orchestrator and programmatic runs leave worktrees in place.",
      "source": "https://docs.mistral.ai/vibe/code/cli/work-with-cli.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/mistral-vibe-cli"
    },
    {
      "slug": "muse-code",
      "name": "Muse Code",
      "company": "Meta Platforms, Inc.",
      "grade": "C",
      "fit": "caveat",
      "why": "The lead can request an isolated Git worktree per subagent, but there is no top-level worktree flag and the compatibility flag does not force isolation.",
      "source": "https://dev.meta.ai/docs/muse-code/extending.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/muse-code"
    },
    {
      "slug": "openai-codex",
      "name": "OpenAI Codex",
      "company": "OpenAI OpCo, LLC",
      "grade": "BB",
      "fit": "caveat",
      "why": "`codex exec --worktree` and `codex fork --worktree` give each run a managed worktree, but exit codes are undocumented and acc-08 hit tool-call limits in cloud containers within 20 minutes.",
      "source": "https://learn.chatgpt.com/docs/developer-commands.md?surface=cli",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/openai-codex"
    },
    {
      "slug": "opencode",
      "name": "OpenCode",
      "company": "Anomaly Innovations, Inc.",
      "grade": "BB",
      "fit": "caveat",
      "why": "A `worktree.directory` setting and server sessions support parallel checkouts, but no run-level worktree flag or status report was found in the CLI pages.",
      "source": "https://opencode.ai/docs/config.md",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/opencode"
    },
    {
      "slug": "openmausbot",
      "name": "OpenMausBot",
      "company": "Supamaus Software Private Limited",
      "grade": "B",
      "fit": "caveat",
      "why": "Bots run on their own computers and `run --json` reports which finished or stalled, but no git worktree or branch isolation is documented.",
      "source": "https://github.com/milind-soni/OpenMausBot/blob/main/docs/headless-cli.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/openmausbot"
    },
    {
      "slug": "orca",
      "name": "Orca",
      "company": "Lovecast Inc.",
      "grade": "C",
      "fit": "caveat",
      "why": "The README claims each agent runs in its own worktree with an orchestrator splitting tasks, but it is the vendor's claim without a documented flag or status report.",
      "source": "https://github.com/stablyai/orca/blob/main/README.md",
      "confidence": "low",
      "url": "https://www.anchorterminal.com/tools/orca"
    },
    {
      "slug": "reflection-mirror",
      "name": "Mirror CLI",
      "company": "Reflection AI",
      "grade": "C",
      "fit": "caveat",
      "why": "`/worktree \u003cname\u003e` gives each agent a `mirror/\u003cname\u003e` branch under `~/.mirror/worktrees/` for parallel tabs, but orchestration is manual in the TUI and `exec` has no worktree flag.",
      "source": "https://developers.reflection.ai/mirror-workspaces.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/reflection-mirror"
    },
    {
      "slug": "rovo-dev-cli",
      "name": "Rovo Dev CLI",
      "company": "Atlassian Pty Ltd",
      "grade": "E",
      "fit": "caveat",
      "why": "`--worktree \u003cname\u003e` isolates each run under `.rovodev/worktrees/`, but an orchestrator learns the outcome only through server mode.",
      "source": "https://support.atlassian.com/rovo/docs/use-worktree-mode-in-rovo-dev-cli/",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/rovo-dev-cli"
    },
    {
      "slug": "unstoppable",
      "name": "Unstoppable",
      "company": "Unstoppable Domains, Inc.",
      "grade": "E",
      "fit": "caveat",
      "why": "Parallel worktrees with review and Create PR are described in the support index, but only index lines and the changelog were read, and chats start in Full access.",
      "source": "https://unstoppable.ai/llms.txt",
      "confidence": "low",
      "url": "https://www.anchorterminal.com/tools/unstoppable"
    },
    {
      "slug": "verdent",
      "name": "Verdent",
      "company": "Verdent AI, Inc.",
      "grade": "E",
      "fit": "caveat",
      "why": "Workspace isolation uses git worktrees per workspace, but the page says manual workspace creation is hidden in the current desktop interface.",
      "source": "https://www.verdent.ai/docs/verdent-manager/advanced-features/workspace-isolation.md",
      "confidence": "low",
      "url": "https://www.anchorterminal.com/tools/verdent"
    },
    {
      "slug": "warp",
      "name": "Warp (Warp Agent CLI and the Oz Agent API)",
      "company": "Denver Technologies, Inc. (d/b/a Warp)",
      "grade": "C",
      "fit": "caveat",
      "why": "Each cloud run gets its own environment and Oz can run Claude Code or Codex, but there is no worktree model for a local checkout.",
      "source": "https://docs.warp.dev/_llms-txt/factories.txt",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/warp"
    },
    {
      "slug": "parallax-agents",
      "name": "Parallax",
      "company": "Conifer Solutions, Inc.",
      "grade": "E",
      "fit": "unchecked",
      "why": "The app runs agents side by side on the owner's subscriptions, but worktree isolation and status reporting are not in the pages read.",
      "confidence": "low",
      "url": "https://www.anchorterminal.com/tools/parallax-agents"
    },
    {
      "slug": "qoder-cli",
      "name": "Qoder CLI",
      "company": "BRIGHT ZENITH PRIVATE LIMITED",
      "grade": "C",
      "fit": "unchecked",
      "why": "docs.qoder.com answered 502 from the fetch proxy on three attempts and a DNS failure through WebFetch, so no page was read.",
      "confidence": "low",
      "url": "https://www.anchorterminal.com/tools/qoder-cli"
    },
    {
      "slug": "qwen-code",
      "name": "Qwen Code",
      "company": "Qwen team, Alibaba Group (no legal entity is named in the repository or the docs)",
      "grade": "BB",
      "fit": "unchecked",
      "why": "Worktrees and Managed Session worktrees pages exist in the navigation but were not read.",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/qwen-code"
    },
    {
      "slug": "agent-zero",
      "name": "Agent Zero",
      "company": "Agent Zero, s.r.o.",
      "grade": "D",
      "fit": "no",
      "why": "No worktree or multi-harness support; Agent Zero is one agent framework with sub-agents.",
      "source": "https://github.com/agent0ai/agent-zero/blob/main/docs/guides/usage.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/agent-zero"
    },
    {
      "slug": "aider",
      "name": "Aider",
      "company": "Aider AI LLC",
      "grade": "D",
      "fit": "no",
      "why": "No worktree flag; parallel runs share the checkout.",
      "source": "https://aider.chat/docs/config/options.html",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/aider"
    },
    {
      "slug": "antigravity-cli",
      "name": "Antigravity CLI",
      "company": "Google LLC",
      "grade": "C",
      "fit": "no",
      "why": "No worktree flag or parallel-session command appears in the headless flag reference or permissions docs; each agy -p runs in the current directory.",
      "source": "https://antigravity.google/docs/cli/headless",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/antigravity-cli"
    },
    {
      "slug": "auggie",
      "name": "Auggie CLI",
      "company": "Augment Computing, Inc.",
      "grade": "D",
      "fit": "no",
      "why": "No worktree flag in the CLI reference; parallel runs would share whatever checkout they are started in.",
      "source": "https://docs.augmentcode.com/cli/reference",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/auggie"
    },
    {
      "slug": "axlr",
      "name": "AXLR",
      "company": "Underpass AI",
      "grade": "BB",
      "fit": "no",
      "why": "No worktree support is documented; isolation comes from per-execution sandbox profiles, not separate checkouts.",
      "source": "https://github.com/underpass-ai/AXLR/blob/main/docs/runtime-contract.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/axlr"
    },
    {
      "slug": "crush",
      "name": "Crush",
      "company": "Charmbracelet, Inc.",
      "grade": "D",
      "fit": "no",
      "why": "No worktree flag; the optional server mode shares workspaces rather than isolating checkouts.",
      "source": "https://github.com/charmbracelet/crush#readme",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/crush"
    },
    {
      "slug": "earendil-pi",
      "name": "Pi",
      "company": "Earendil Inc.",
      "grade": "B",
      "fit": "no",
      "why": "No worktree flag; parallel runs share the checkout they start in.",
      "source": "https://github.com/earendil-works/pi/blob/main/packages/coding-agent/docs/cli.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/earendil-pi"
    },
    {
      "slug": "goose",
      "name": "goose",
      "company": "Agentic AI Foundation (Linux Foundation)",
      "grade": "BB",
      "fit": "no",
      "why": "No worktree flag; sub-recipes run in the same directory.",
      "source": "https://github.com/aaif-goose/goose/blob/main/documentation/docs/guides/goose-cli-commands.md",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/goose"
    },
    {
      "slug": "harness2",
      "name": "Harness²",
      "company": "Google LLC",
      "grade": "E",
      "fit": "no",
      "why": "--workers runs benchmark domains concurrently; no worktrees or hand-offs between harnesses on a repository.",
      "source": "https://github.com/google-research/harness2/blob/main/docs/EXPERIMENTS.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/harness2"
    },
    {
      "slug": "jules",
      "name": "Jules",
      "company": "Google LLC",
      "grade": "D",
      "fit": "no",
      "why": "jules remote new --parallel started sessions in separate VMs, but new sessions stopped after 11 November 2026.",
      "source": "https://jules.google/docs/cli/reference",
      "confidence": "high",
      "url": "https://www.anchorterminal.com/tools/jules"
    },
    {
      "slug": "junie",
      "name": "Junie CLI",
      "company": "JetBrains s.r.o.",
      "grade": "D",
      "fit": "no",
      "why": "No worktree flag in the CLI reference.",
      "source": "https://junie.jetbrains.com/docs/parameters.html",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/junie"
    },
    {
      "slug": "kimi-code-cli",
      "name": "Kimi Code CLI",
      "company": "Moonshot AI (the global Kimi terms name NOVASCENT PRIVATE LIMITED, Singapore)",
      "grade": "C",
      "fit": "no",
      "why": "No `--worktree` flag or worktree docs; remote environments and subagents exist but share the session's checkout.",
      "source": "https://moonshotai.github.io/kimi-code/en/reference/kimi-command",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/kimi-code-cli"
    },
    {
      "slug": "kiro-cli",
      "name": "Kiro CLI",
      "company": "Amazon Web Services, Inc.",
      "grade": "B",
      "fit": "no",
      "why": "No worktree flag in the command reference; runs share the current checkout.",
      "source": "https://kiro.dev/docs/reference/cli-commands/",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/kiro-cli"
    },
    {
      "slug": "letta",
      "name": "Letta",
      "company": "Letta",
      "grade": "B",
      "fit": "no",
      "why": "No worktree or branch isolation is documented; subagents run inside the same agent.",
      "source": "https://docs.letta.com/platform/cli/reference/index.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/letta"
    },
    {
      "slug": "mini-swe-agent",
      "name": "mini-swe-agent",
      "company": "No legal entity named. The MIT licence names Kilian A. Lieret and Carlos E. Jimenez",
      "grade": "B",
      "fit": "no",
      "why": "No worktree support; the local environment runs in the current directory.",
      "source": "https://mini-swe-agent.com/latest/advanced/environments/",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/mini-swe-agent"
    },
    {
      "slug": "minimax-code",
      "name": "MiniMax Code",
      "company": "Nanonoble Pte. Ltd.",
      "grade": "B",
      "fit": "no",
      "why": "No worktree flag among the `exec` options; `--cwd` picks an existing directory.",
      "source": "https://agent.minimax.io/docs/cli/automation.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/minimax-code"
    },
    {
      "slug": "openai-agents-api",
      "name": "OpenAI Agents API",
      "company": "OpenAI OpCo, LLC",
      "grade": "BB",
      "fit": "no",
      "why": "Sessions get isolated sandboxes, not branches of one checkout, and no hand-off to other harnesses exists.",
      "source": "https://developers.openai.com/api/docs/guides/agents-api/environments/openai-hosted.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/openai-agents-api"
    },
    {
      "slug": "openhands",
      "name": "OpenHands",
      "company": "All Hands AI",
      "grade": "BB",
      "fit": "no",
      "why": "Conversations get their own Docker sandbox, not worktrees of one checkout, and no hand-off between harnesses is documented.",
      "source": "https://docs.openhands.dev/llms.txt",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/openhands"
    },
    {
      "slug": "prime-agent",
      "name": "Prime Agent",
      "company": "Prime Intellect, Inc.",
      "grade": "C",
      "fit": "no",
      "why": "`rlm.spawn` subagents run in parallel inside one session; no worktree isolation is documented.",
      "source": "https://github.com/PrimeIntellect-ai/prime-agent/blob/main/README.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/prime-agent"
    },
    {
      "slug": "vibe-trading",
      "name": "Vibe-Trading",
      "company": "HKUDS",
      "grade": "C",
      "fit": "no",
      "why": "No worktrees or coding agents.",
      "source": "https://github.com/HKUDS/Vibe-Trading/blob/main/README.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/vibe-trading"
    }
  ],
  "marks": "yes: the docs show it does the job as the success test says. caveat: it does the job with a limit an agent should know. no: it doesn't do this job. unchecked: we couldn't read it.",
  "meta": {
    "attribution": "Anchor Terminal (https://www.anchorterminal.com)",
    "attributionRequired": true,
    "credit": "CC BY 4.0, credit Anchor Terminal and link the page or API address you took it from (https://www.anchorterminal.com/terms/#data)",
    "docs": "https://www.anchorterminal.com/docs/",
    "generatedAt": "2026-11-16",
    "license": "CC-BY-4.0",
    "method": "https://www.anchorterminal.com/benchmark/",
    "methodology": "0.8",
    "openapi": "https://www.anchorterminal.com/openapi.json",
    "preview": false,
    "run": "2026-10-01",
    "runLabel": "October 2026 research run",
    "terms": "https://www.anchorterminal.com/terms/#data"
  },
  "useCase": {
    "accounts": 2,
    "after": [
      "code-hosting.open-pull-request"
    ],
    "area": "frameworks",
    "asks": [
      "run three tasks in parallel on separate branches",
      "give this to Codex and that to Claude Code",
      "start a worker for each issue"
    ],
    "before": [
      "issue-tracking.list-issues"
    ],
    "blockers": [
      "orchestrator runs every child with permissions bypassed and no sandbox",
      "tool-calling limits in cloud containers within 20 minutes (acc-08)",
      "child harness shows a dialog the orchestrator cannot answer (acc-15)"
    ],
    "capabilities": [
      "agent.multi-agent",
      "agent.harness"
    ],
    "category": "agent-harnesses",
    "categoryName": "Agent harnesses",
    "checked": "2026-11-12",
    "companies": {
      "fit": 10,
      "fitWithLimit": 23,
      "tagged": 77
    },
    "competition": "crowded",
    "fits": [
      "claude-code",
      "cursor-cli",
      "devin",
      "droid-cli",
      "emdash",
      "paperclip",
      "paseo",
      "sculptor",
      "t3-code",
      "traycer"
    ],
    "fitsWithLimit": [
      "amp",
      "clawboo",
      "cli-manager",
      "cline",
      "conductor",
      "deepseek-harness",
      "deepseek-superagent",
      "gemini-cli",
      "github-copilot-cli",
      "harness-autonomous",
      "harnessrouter",
      "kilo-code-cli",
      "mistral-vibe-cli",
      "muse-code",
      "openai-codex",
      "opencode",
      "openmausbot",
      "orca",
      "reflection-mirror",
      "rovo-dev-cli",
      "unstoppable",
      "verdent",
      "warp"
    ],
    "id": "agent-harnesses.parallel-worktrees",
    "listings": {
      "caveat": 23,
      "no": 20,
      "notRead": 0,
      "tagged": 88,
      "unchecked": 3,
      "yes": 10
    },
    "name": "Run several agents at once in separate worktrees or hand tasks between harnesses",
    "status": "draft",
    "success": "Each task ends on its own branch with a reviewable diff, no two agents touched the same checkout, and the orchestrator can tell which finished and which stalled.",
    "url": "https://www.anchorterminal.com/api/v1/use-cases/agent-harnesses.parallel-worktrees.json",
    "weight": 4,
    "weightBasis": "2 of 33 accounts (acc-08, acc-15); 15 of 56 listings are orchestrators, so the weight is low on evidence and should rise when accounts appear"
  }
}
