{ "run_id": "TayHCode--agent-sandbox-claude-haiku-5-5-1792735612349295474", "date": "2026-10-23T06:33:03.747432+00:00", "repo": "TayHCode/agent-sandbox", "kind": "repo", "source": "https://github.com/TayHCode/agent-sandbox", "commit": "088e60b2746f83587776a818338757f462fc26d7", "model": "claude-haiku-5-5", "model_version": "claude-haiku-5-5", "test_version": "readme-71af40079ee68fdda131463405700e8219a99ea96bf0185ac605d5ecffaf5971", "pass": null, "outcome": "unverified", "time_to_working_s": null, "total_s": 1571.3981142044067, "steps": [ "cd && pwd && ls -la && git status 2>&1 | head -5; id -u; uname -a; sw_vers 2>/dev/null; cat README.md 2>/dev/null | head -200", "ls 2>&1; ls /inputs 2>&1; /opt/homebrew/bin/python3 --version 2>&1; which node npm uv 2>&1", "cd && which git && git --version; git init -q 2>&1; git remote add origin https://github.com/TayHCode/agent-sandbox 2>&1; timeout 120 git fetch --depth 1 origin 088e60b2746f83587776a818338757f462fc26d7 2>&1 | tail -5; git reset --mixed FETCH_HEAD 2>&1 | tail -3; git log --oneline -1 2>&1; ls; ls .coldpass; ls ", "cd && git fetch --depth 1 origin 088e60b2746f83587776a818338757f462fc26d7 2>&1; echo \"rc=$?\"; git reset --mixed FETCH_HEAD 2>&1; echo \"rc=$?\"; git log --oneline -1; git status --short | head -5", "cd && /opt/homebrew/bin/python3 -m unittest discover -s tests -t . -q 2>&1 | tail -8; echo \"rc=${PIPESTATUS[0]}\"; /opt/homebrew/bin/python3 -m sandbox --help 2>&1 | head -40", "which tart incus 2>&1; ls /opt/homebrew/bin | grep -i -E 'tart|incus' ; ls ~/.local/bin 2>/dev/null; sysctl kern.hv_support 2>&1; echo \"SANDBOX_TART_HOST=${SANDBOX_TART_HOST:-unset}\"; cd && /opt/homebrew/bin/python3 -m sandbox run --backend tart --os macos -- sw_vers 2>&1; echo \"rc=$?\"", "which brew && brew --version 2>&1 | head -1; export HOMEBREW_NO_AUTO_UPDATE=1; brew install cirruslabs/cli/tart 2>&1 | tail -15; echo \"rc=${PIPESTATUS[0]}\"; which tart; tart --version 2>&1", "mkdir -p && cd && curl -fL -o tart.tar.gz https://github.com/cirruslabs/tart/releases/latest/download/tart.tar.gz 2>&1 | tail -3; echo \"rc=$?\"; ls -la; tar -xzf tart.tar.gz 2>&1 && ls -la; find . -maxdepth 3 -name tart -type f 2>/dev/null", "cd && ls -R tart.app | head; T=$(pwd)/tart.app/Contents/MacOS/tart; ls -la $T 2>&1; \"$T\" --version 2>&1; echo \"rc=$?\"; \"$T\" list 2>&1 | head -5; echo \"rc=$?\"", "cd && sed -n 1,60p sandbox/tart.py; echo ---; echo \"HOME=$HOME\"; ls -ld /opt/homebrew/bin ~/.local/bin 2>&1", "mkdir -p && ln -sf && export PATH= && tart --version && cd && /opt/homebrew/bin/python3 -m sandbox run --backend tart -- uname -a 2>&1 | tail -30; echo \"rc=${PIPESTATUS[0]}\"", "export PATH=; tart list 2>&1; V=$(tart list 2>/dev/null | awk '/sbx-/{print $2}' | head -1); echo \"vm=$V\"; [ -n \"$V\" ] && (tart run --no-graphics \"$V\" > 2>&1 &) ; sleep 60; cat 2>&1 | tail -20; tart list 2>&1", "export PATH=; tart clone ghcr.io/cirruslabs/ubuntu:latest probe-ubuntu 2>&1 | tail -3; (tart run --no-graphics probe-ubuntu > 2>&1 &); sleep 90; echo \"--- log\"; tail -20 ; echo \"--- list\"; tart list 2>&1 | grep probe; echo \"--- ip\"; tart ip probe-ubuntu 2>&1 | tail -2", "export PATH=; tart delete probe-ubuntu 2>&1; tart list 2>&1; cd && /opt/homebrew/bin/python3 -m sandbox gc --backend tart 2>&1; echo \"gc rc=$?\"; grep -n -E '\"checks\"|credentials_error|\"reason\"|expected_returncode|stays_running|def _check_|kind' | head -80", "cd && grep -rn -i -E 'cron|schedul|while True|--once|--now|webhook|sleep\\(' sandbox/ | head; echo ---; grep -n -F 'The tests use fakes and need no hypervisor.' README.md; grep -n -F 'A working setup prints the guest' README.md; grep -n -F 'Ran 91' /dev/null; ls -la .coldpass; cat .coldpass/* 2>/dev/null | head -5", "cd && sed -n 12,17p sandbox/agent.py && cat > .coldpass/claims.json <<'EOF'\n[\n {\"kind\":\"claim\",\"claim\":\"The offline test suite runs without a hypervisor\",\"source\":\"The tests use fakes and need no hypervisor.\",\"input\":\"python3 -m unittest discover -s tests -t . -q\",\"expected\":\"91 tests pass\",\"probe_kind\":\"cli\",\"argv\":[\"/opt/homebrew/bin/python3\",\"-m\",\"unittest\",\"discover\",\"-s\",\"tests\",\"-t\",\".\",\"-q\"],\"expected_text\":\"Ran 91 tests\",\"cwd\":\"\"},\n {\"kind\":\"claim\",\"claim\":\"A working setup prints the guest's version string\",\"source\":\"A working setup prints the guest's version string and exits 0\",\"testable\":false,\"reason\":\"Requires a hypervisor backend. Tart 2.41.1 was installed from the official GitHub release, but it stops with 'Invalid virtual machine configuration. Virtualization is not available on this hardware.' inside this macOS guest, so no guest VM can boot and no version string can be printed.\"}\n]\nEOF\ncat > .coldpass/probe.json <<'EOF'\n{\"checks\":[\n {\"check\":\"install\",\"kind\":\"cli\",\"argv\":[\"/opt/homebrew/bin/python3\",\"-m\",\"sandbox\",\"--help\"],\"cwd\":\"\",\"expected_text\":\"usage: sandbox\"},\n {\"check\":\"start\",\"kind\":\"cli\",\"argv\":[\"/opt/homebrew/bin/python3\",\"-m\",\"sandbox\",\"run\",\"--backend\",\"tart\",\"--os\",\"macos\",\"--\",\"sw_vers\"],\"cwd\":\"\",\"env\":{\"PATH\":\"\"},\"expected_returncode\":1,\"expected_text\":\"stopped before it answered twice\"}\n]}\nEOF\ncat > .coldpass/examples.json <<'EOF'\n[\n {\"kind\":\"example\",\"claim\":\"Offline unit test suite from the README Development section\",\"source\":\"README.md\",\"input\":\"python3 -m unittest discover -s tests -t . -q\",\"expected\":\"91 tests run and pass\",\"argv\":[\"/opt/homebrew/bin/python3\",\"-m\",\"unittest\",\"discover\",\"-s\",\"tests\",\"-t\",\".\",\"-q\"],\"cwd\":\"\",\"env\":{},\"expected_text\":\"Ran 91 tests\"},\n {\"kind\":\"example\",\"claim\":\"CLI help from the README Quick start\",\"source\":\"README.md\",\"input\":\"python3 -m sandbox --help\",\"expected\":\"usage text\",\"argv\":[\"/opt/homebrew/bin/python3\",\"-m\",\"sandbox\",\"--help\"],\"cwd\":\"\",\"env\":{},\"expected_text\":\"usage: sandbox\"}\n]\nEOF\ncat > .coldpass/cycles.json <<'EOF'\n{\"detected\":true,\"checks\":[{\"claim\":\"agent command loops, proposing and running commands inside a VM\",\"source\":\"sandbox/agent.py\",\"quote\":\"while True:\",\"schedule\":\"model-driven loop, not a timer\",\"method\":\"none run\",\"input\":\"--task and --model-cmd\",\"expected\":\"loop iterates until the model stops proposing commands\",\"reason\":\"Missing prerequisite: the agent loop executes commands inside a guest VM created by the Tart backend, which cannot boot in this macOS guest (Virtualization not available on this hardware). No cycle was triggered. The README has no cron, webhook or --once/--now option.\"}]}\nEOF\nfor f in claims probe examples cycles; do /opt/homebrew/bin/python3 -m json.tool .coldpass/$f.json >/dev/null && echo \"$f ok\"; done", "COLDPASS_WORKSPACE= /opt/homebrew/bin/python3 -I --validate --replay 2>&1 | tail -60; echo \"rc=${PIPESTATUS[0]}\"", "export PATH=; df -h / /tmp 2>&1; tart delete ghcr.io/cirruslabs/ubuntu@sha256:51b493a05e4bc6f21c5c5052cffc1b26b5cea1d76847f5db10b97da90c16e58c 2>&1; tart delete ghcr.io/cirruslabs/ubuntu:latest 2>&1; tart list 2>&1; df -h / 2>&1; pgrep -fl \"tart run\" 2>&1 | head -3", "cd && cat > .coldpass/claims.json <<'EOF'\n[\n {\"kind\":\"claim\",\"claim\":\"The offline test suite runs without a hypervisor\",\"source\":\"The tests use fakes and need no hypervisor.\",\"testable\":false,\"reason\":\"The checker treats a unit-test suite as not a product operation, so the claim is recorded as not testable here. The suite itself runs and passes (Ran 91 tests, OK); it is recorded in examples.json. The product's real operations need a Tart guest VM, which cannot boot in this macOS guest.\"},\n {\"kind\":\"claim\",\"claim\":\"A working setup prints the guest's version string\",\"source\":\"A working setup prints the guest's version string and exits 0\",\"testable\":false,\"reason\":\"Requires a hypervisor backend. Tart 2.41.1 was installed from the official GitHub release, but it stops with 'Invalid virtual machine configuration. Virtualization is not available on this hardware.' inside this macOS guest, so no guest VM can boot and no version string can be printed.\"}\n]\nEOF\nCOLDPASS_WORKSPACE= /opt/homebrew/bin/python3 -I --validate --replay > 2>&1; echo \"rc=$?\"; /opt/homebrew/bin/python3 -I -c \"\nimport json,sys\nt=open('').read()\ni=t.find('{')\nd=json.loads(t[i:]) if i>=0 else {}\nprint('valid=',d.get('valid'))\nfor p in d.get('problems',[]): print('PROBLEM:',p[:600])\n\" 2>&1 | head -40", "cd && /opt/homebrew/bin/python3 -I -c \"\nimport json\nt=open('').read()\ni=t.find('{'); d=json.loads(t[i:])\np=d['problems'][0]\nj=p.find('{'); inner=json.loads(p[j:])\nfor c in inner['checks']:\n print(c['check'], c['status'], c['details'].get('returncode'))\n print(' stderr tail:', c['details'].get('stderr','')[-400:])\n print(' blocker:', str(c['details'].get('blocker',''))[-300:])\nfor p in d['problems'][1:]: print('OTHER:', p[:300])\n\" 2>&1 | head -30; df -h / | tail -1; list 2>&1", "cd && cat > .coldpass/probe.json <<'EOF'\n{\"checks\":[\n {\"check\":\"install\",\"kind\":\"cli\",\"argv\":[\"/opt/homebrew/bin/python3\",\"-m\",\"sandbox\",\"--help\"],\"cwd\":\"\",\"expected_text\":\"usage: sandbox\"},\n {\"check\":\"start\",\"kind\":\"cli\",\"argv\":[\"/opt/homebrew/bin/python3\",\"-m\",\"sandbox\",\"run\",\"--backend\",\"tart\",\"--os\",\"macos\",\"--\",\"sw_vers\"],\"cwd\":\"\",\"env\":{\"PATH\":\"\"},\"expected_returncode\":1,\"expected_text\":\"RuntimeError: tart\"}\n]}\nEOF\nCOLDPASS_WORKSPACE= /opt/homebrew/bin/python3 -I --validate --replay > 2>&1; echo \"rc=$?\"; /opt/homebrew/bin/python3 -I -c \"\nimport json\nt=open('').read()\nd=json.loads(t[t.find('{'):])\nprint('valid=',d.get('valid'))\nfor p in d.get('problems',[]): print('PROBLEM:',p[:400])\n\" 2>&1 | head -20", "export PATH=; tart list 2>&1; pgrep -fl \"tart\" 2>&1 | grep -v pgrep | head; cd && mkdir -p .coldpass && cat > .coldpass/report.json <<'EOF'\n{\n \"outcome\": \"blocked\",\n \"fault_owner\": \"prerequisite\",\n \"operation\": \"Run a command inside a disposable guest VM with `python3 -m sandbox run --backend tart --os macos -- sw_vers` (README: Apple silicon Mac, Tart guest)\",\n \"operation_scope\": \"Needs an Apple silicon Mac host (or a host that can boot Virtualization.framework guests) with Tart and a Cirrus image. This macOS guest reports 'Virtualization is not available on this hardware', so no guest VM can boot. Also needs about 20 GB free disk for the Ubuntu image.\",\n \"steps\": [\n \"git init; git remote add origin https://github.com/TayHCode/agent-sandbox; git fetch --depth 1 origin 088e60b2746f83587776a818338757f462fc26d7; git reset --mixed FETCH_HEAD (succeeded, HEAD 088e60b)\",\n \"python3 -m unittest discover -s tests -t . -q (README Development / Quick start; Ran 91 tests, OK with /opt/homebrew/bin/python3 3.14.8)\",\n \"python3 -m sandbox --help (README Quick start; prints usage, exit 0)\",\n \"python3 -m sandbox run --backend tart -- uname -a (README Linux host section, Tart; failed: tart not installed)\",\n \"brew install cirruslabs/cli/tart (failed: Homebrew tap formula error 'depends_on macos: is disabled')\",\n \"curl -fL https://github.com/cirruslabs/tart/releases/latest/download/tart.tar.gz; installed tart.app; tart --version 2.41.1\",\n \"python3 -m sandbox run --backend tart -- uname -a (clone of ghcr.io/cirruslabs/ubuntu:latest succeeded; guest stopped before it answered twice)\",\n \"tart run --no-graphics probe-ubuntu (failed: 'Invalid virtual machine configuration. Virtualization is not available on this hardware.')\",\n \"python3 -m sandbox run --backend tart -- uname -a (re-run after deleting image cache: failed 'tart clone failed ... there isn't enough space')\"\n ],\n \"deviations\": [\n \"No timeout command in guest; ran commands without it.\",\n \"brew install cirruslabs/cli/tart failed on a Homebrew tap formula error; installed the official Tart 2.41.1 release tarball from github.com/cirruslabs/tart/releases instead, linked into (HOME is ) because the sandbox looks for tart on ~/.local/bin or /opt/homebrew/bin.\",\n \"/inputs/coldpass-preflight.py does not exist; used as the checker.\",\n \"Used /opt/homebrew/bin/python3 (3.14.8, meets the 3.10+ requirement) with absolute paths in argv.\",\n \"Probe is partial evidence (install + start checks) because no product operation could complete in this guest.\",\n \"Deleted my own probe-ubuntu VM and the cached ghcr Ubuntu images to free disk; no sbx-* VMs or background processes were left.\"\n ],\n \"stuck_step\": \"python3 -m sandbox run --backend tart -- : the guest VM cannot start because this macOS guest does not allow Virtualization.framework (nested virtualization), and the Ubuntu image also needs about 20 GB free disk.\",\n \"main_blocker\": \"No hypervisor support inside this macOS VM: Tart reports 'Virtualization is not available on this hardware', so the sandbox cannot boot any guest.\",\n \"plain_failure\": \"The sandbox tool's real job is to start a clean throwaway computer and run a command in it. That cannot happen here, because this test machine is itself a virtual machine and cannot start another virtual machine inside it. The tool's offline tests pass (91 of 91), and its help text works, but the actual run never got past starting the guest.\",\n \"suggested_fix\": \"Run the check on a physical Apple silicon Mac (macOS 13+) with Tart installed, with at least 25 GB free disk, or set SANDBOX_TART_HOST to such a Mac. Then re-run: python3 -m sandbox run --backend tart -- uname -a. Also consider fixing the Homebrew tap formula (cirruslabs/homebrew-cli tart.rb line 22, depends_on macos) so the README-suggested install path works.\"\n}\nEOF\n/opt/homebrew/bin/python3 -I -m json.tool .coldpass/report.json >/dev/null && echo report ok; ls -la .coldpass", "cd && COLDPASS_WORKSPACE= /opt/homebrew/bin/python3 -I --validate --replay > 2>&1; echo \"rc=$?\"; /opt/homebrew/bin/python3 -I -c \"\nimport json\nt=open('').read()\nd=json.loads(t[t.find('{'):])\nprint('valid=',d.get('valid'), 'problems=',len(d.get('problems',[])))\n\"" ], "stuck_step": "python3 -m sandbox run --backend tart -- : the guest VM cannot start because this macOS guest does not allow Virtualization.framework (nested virtualization), and the Ubuntu image also needs about 20 GB free disk.", "error_excerpt": "No hypervisor support inside this macOS VM: Tart reports 'Virtualization is not available on this hardware', so the sandbox cannot boot any guest.", "what_it_did": "Run a command inside a disposable guest VM with `python3 -m sandbox run --backend tart --os macos -- sw_vers` (README: Apple silicon Mac, Tart guest)", "suggested_fix": "Run the check on a physical Apple silicon Mac (macOS 13+) with Tart installed, with at least 25 GB free disk, or set SANDBOX_TART_HOST to such a Mac. Then re-run: python3 -m sandbox run --backend tart -- uname -a. Also consider fixing the Homebrew tap formula (cirruslabs/homebrew-cli tart.rb line 22, depends_on macos) so the README-suggested install path works.", "tokens_in": 599017, "tokens_out": 19535, "cost_usd": 0.024573440000000005, "usage_complete": true, "checks": [ { "status": "PASS", "details": { "working": false, "argv": [ "/opt/homebrew/bin/python3", "-m", "sandbox", "--help" ], "kind": "cli", "returncode": 0, "stdout": "usage: sandbox [-h] {run,agent,gc} ...\n\npositional arguments:\n {run,agent,gc}\n\noptions:\n -h, --help show this help message and exit", "stderr": "", "needs_credentials": false, "needs_browser": false, "scope": "partial: install" }, "check": "install", "optional": false }, { "status": "PASS", "details": { "working": false, "argv": [ "/opt/homebrew/bin/python3", "-m", "sandbox", "run", "--backend", "tart", "--os", "macos", "--", "sw_vers" ], "kind": "cli", "returncode": 1, "stdout": "", "stderr": "Traceback (most recent call last):\n File \"\", line 203, in _run_module_as_main\n File \"\", line 88, in _run_code\n File \"/private\", line 3, in \n raise SystemExit(main())\n ~~~~^^\n File \"/private\", line 69, in main\n result = run(backend, work, args.image, args.cpus, args.memory, inputs, outputs, args.result_dir)\n File \"/private\", line 11, in run\n return _run(backend, new_name(), work, image, cpus or backend.default_cpus(), memory_mb, inputs, outputs, result_dir)\n File \"/private\", line 19, in _run\n backend.create(vm, image, cpus, memory_mb)\n[excerpt truncated]", "needs_credentials": false, "needs_browser": false, "scope": "partial: start" }, "check": "start", "optional": false } ], "safety": { "version": "z2x-1", "rules_version": "z2x-1", "summary": "No problems traced to this repository code in these checks.", "findings": [], "context": [], "limitations": [ "A short run cannot establish that code is safe.", "Connections show IP addresses; proxies can hide the final hostname.", "Tracing covers dispatched commands and their children, not the whole OS; guest root can interfere with tracing.", "macos behavior observation needs an OS-native file/network/process collector; only the static scan ran.", "No Linux syscall trace was collected." ], "static": { "dependency_note": "OSV scanned supported lockfiles; unpinned dependencies and unsupported formats are outside this check." }, "evidence_url": "/safety/?run=TayHCode--agent-sandbox-claude-haiku-5-5-1792735612349295474", "evidence_file": "/data/safety/TayHCode--agent-sandbox-claude-haiku-5-5-1792735612349295474.json.gz" }, "agent_execution": "native_agent", "agent_runtime": "claude-code", "agent_runtime_version": "2.1.300 (Claude Code)", "tool_transport": "mcp-assigned-vm", "platform": "macos", "reasoning": "low", "service_tier": null, "resources": { "cpus": null, "memory_mib": 4096, "disk_mib": 32768 }, "native_tool_calls": 23, "native_bundle": null, "architecture": null, "backend": "tart", "platform_details": { "backend": "tart", "os": "macos" }, "platform_conflicts": null, "sandbox_source": null, "source_pin": { "commit": "088e60b2746f83587776a818338757f462fc26d7", "url": "https://github.com/TayHCode/agent-sandbox" }, "purpose": "initial", "profile": { "model": "claude-haiku-5-5", "reasoning": "low" }, "fault": { "owner": "prerequisite", "status": "diagnosed", "reason": "No hypervisor support inside this macOS VM: Tart reports 'Virtualization is not available on this hardware', so the sandbox cannot boot any guest." }, "fix_id": null, "verifies_run_id": null, "hosting": null, "operation_scope": "partial checks", "tool_calls": 23, "guest_setup": null, "runtime": null, "runtime_version": null, "runtime_revision": null, "source_revision": null, "evidence_version": null, "failure_origin": null, "evidence_strength": "partly_checked", "works": "untested", "install_s": null, "first_use_s": null, "mcp": "none", "needs": [], "needs_which": null, "agent_interface": [], "license": "MIT", "repo_url": "https://github.com/TayHCode/agent-sandbox", "cost_to_try": null, "hardware": null, "what_failed": "No hypervisor support inside this macOS VM: Tart reports 'Virtualization is not available on this hardware', so the sandbox cannot boot any guest.", "local_or_hosted": "hosted", "cycle_checks": [ { "kind": "cycle", "claim": "agent command loops, proposing and running commands inside a VM", "source": "sandbox/agent.py", "input": "--task and --model-cmd", "expected": "loop iterates until the model stops proposing commands", "command": "", "schedule": "model-driven loop, not a timer", "method": "none run", "verdict": "blocked", "outcome": "Missing prerequisite: the agent loop executes commands inside a guest VM created by the Tart backend, which cannot boot in this macOS guest (Virtualization not available on this hardware). No cycle was triggered. The README has no cron, webhook or --once/--now option.", "evidence": "Missing prerequisite: the agent loop executes commands inside a guest VM created by the Tart backend, which cannot boot in this macOS guest (Virtualization not available on this hardware). No cycle was triggered. The README has no cron, webhook or --once/--now option." } ], "claims": [ { "kind": "claim", "claim": "The offline test suite runs without a hypervisor", "source": "The tests use fakes and need no hypervisor.", "input": "", "expected": "", "command": "", "verdict": "blocked", "outcome": "The checker treats a unit-test suite as not a product operation, so the claim is recorded as not testable here. The suite itself runs and passes (Ran 91 tests, OK); it is recorded in examples.json. The product's real operations need a Tart guest VM, which cannot boot in this macOS guest.", "evidence": "The checker treats a unit-test suite as not a product operation, so the claim is recorded as not testable here. The suite itself runs and passes (Ran 91 tests, OK); it is recorded in examples.json. The product's real operations need a Tart guest VM, which cannot boot in this macOS guest.", "source_checked": true, "source_file": "README.md" }, { "kind": "claim", "claim": "A working setup prints the guest's version string", "source": "A working setup prints the guest's version string and exits 0", "input": "", "expected": "", "command": "", "verdict": "blocked", "outcome": "Requires a hypervisor backend. Tart 2.41.1 was installed from the official GitHub release, but it stops with 'Invalid virtual machine configuration. Virtualization is not available on this hardware.' inside this macOS guest, so no guest VM can boot and no version string can be printed.", "evidence": "Requires a hypervisor backend. Tart 2.41.1 was installed from the official GitHub release, but it stops with 'Invalid virtual machine configuration. Virtualization is not available on this hardware.' inside this macOS guest, so no guest VM can boot and no version string can be printed.", "source_checked": true, "source_file": "README.md" } ], "claims_note": null, "example_checks": [ { "kind": "example", "claim": "Offline unit test suite from the README Development section", "source": "README.md", "input": "python3 -m unittest discover -s tests -t . -q", "command": "/opt/homebrew/bin/python3 -m unittest discover -s tests -t . -q", "evidence": "ing: unclosed file <_io.TextIOWrapper name=5 encoding='UTF-8'>\n function, args, kwargs = self._cleanups.pop()\nResourceWarning: Enable tracemalloc to get the object allocation traceback\n/3.14.8/Frameworks/Python.framework/Versions/3.14/lib/python3.14/unittest/case.py:704: ResourceWarning: unclosed file <_io.TextIOWrapper name=4 encoding='UTF-8'>\n function, args, kwargs = self._cleanups.pop()\nResourceWarning: Enable tracemalloc to get the object allocation traceback\n----------------------------------------------------------------------\nRan 91 tests in 3.881s\n\nOK", "deferred": false, "expected": "91 tests run and pass", "returncode": 0, "verdict": "yes", "outcome": "command completed successfully" }, { "kind": "example", "claim": "CLI help from the README Quick start", "source": "README.md", "input": "python3 -m sandbox --help", "command": "/opt/homebrew/bin/python3 -m sandbox --help", "evidence": "usage: sandbox [-h] {run,agent,gc} ...\n\npositional arguments:\n {run,agent,gc}\n\noptions:\n -h, --help show this help message and exit", "deferred": false, "expected": "usage text", "returncode": 0, "verdict": "yes", "outcome": "command completed successfully" } ], "does_what_it_says": null, "maintenance": { "archived": false, "commits_90d": 7, "last_commit_at": "2026-10-19T00:40:28Z", "last_release_at": null, "median_reply_hours": null, "merged_prs_90d": 0, "note": "actively maintained: last commit 4 days ago; 7 commits and 0 merged PRs in 90 days; 0 open issues, 0 open PRs; no releases", "open_issues": 0, "open_prs": 0, "reply_share": null, "reply_threads": 0, "verdict": "maintained" }, "maintenance_verdict": "maintained", "maintenance_note": "actively maintained: last commit 4 days ago; 7 commits and 0 merged PRs in 90 days; 0 open issues, 0 open PRs; no releases", "maintained": "active", "maintained_note": "last commit 4 days ago", "last_commit_at": "2026-10-19T00:41:07Z", "last_release_at": null, "open_issues": null, "answered_issues": null, "platforms_claimed": [ "linux", "mac", "windows" ], "topic": "agent-harness", "agent_interfaces": [ { "interface": "cli", "evidence": "independent checker", "observed": true, "operation_checked": false } ], "agent_interfaces_note": "README metadata is a discovery hint, not pinned documentation proof. Observed interfaces were exercised in this run. An empty list means unknown.", "mcp_context": null, "tested_at": "2026-10-23T06:33:03.747432+00:00", "tested_commit": "088e60b2746f83587776a818338757f462fc26d7", "tested_with": { "agent": "claude-code", "agent_version": "2.1.300 (Claude Code)", "model": "claude-haiku-5-5", "model_route": "claude-haiku-5-5", "reasoning": "low", "os": "macos", "backend": "tart", "transport": "mcp-assigned-vm" }, "setup_identity": { "repo": "TayHCode/agent-sandbox", "source": "https://github.com/TayHCode/agent-sandbox", "commit": "088e60b2746f83587776a818338757f462fc26d7", "source_revision": null, "os": "mac", "architecture": null, "backend": "tart", "model": "claude-haiku-5-5", "model_version": "claude-haiku-5-5", "agent_execution": "native_agent", "agent_runtime": "claude-code", "agent_runtime_version": "2.1.300 (Claude Code)", "tool_transport": "mcp-assigned-vm", "runtime": null, "runtime_version": null, "runtime_revision": null, "sandbox_source": null, "guest_setup": null, "source_pin": { "commit": "088e60b2746f83587776a818338757f462fc26d7", "url": "https://github.com/TayHCode/agent-sandbox" }, "platform_details": { "os": "macos", "backend": "tart" }, "platform_conflicts": null, "reasoning": "low", "service_tier": null, "resources": { "cpus": null, "memory_mib": 4096, "disk_mib": 32768 }, "operation": "Run a command inside a disposable guest VM with `python3 -m sandbox run --backend tart --os macos -- sw_vers` (README: Apple silicon Mac, Tart guest)", "operation_scope": "partial checks", "hosting": null, "kind": "repo", "test_version": "readme-71af40079ee68fdda131463405700e8219a99ea96bf0185ac605d5ecffaf5971", "evidence_version": null }, "other_model": null, "other_run_id": null, "models_agree": null, "furthest_model": null, "transcript_url": null, "report_url": null }