coldpass

Arekejoker/qwen21-nsfw-pose-skill

GitHub ↗ · How this check works

Per-setup public labels: Linux / Haiku: Works

Static status badge · SVG · read-only JSON. Embed with ![Coldpass status](https://coldpass.dev/badge/Arekejoker--qwen21-nsfw-pose-skill.svg).

agent-skillcomfyuiimage-generationopenclawpose-controlqwen-image

Works

Only older checks are saved; no current product test.

Answers apply to each recorded setup, revision and operation. Earlier observations remain in Run details.

Tested on Linux
PlatformHaikuLuna
Linuxyes

Linux / Haiku 5.5: yes · Oct 18, 2026 · eca9d7fbf8c5 · claude-code / 2.1.300 (Claude Code)
Recorded operation: /usr/bin/python3 scripts/validate_package.py

Needs keys or browser
None seen
Issue
none found in recorded check

Safety: no observations were recorded for the selected revision. Functional results do not establish a safety result.

Run details

Haiku 5.5

Native agent · Checked operation passed

Checked operation /usr/bin/python3 scripts/validate_package.py

Agent-reported scope: Ran the CI workflow's documented offline check: python scripts/validate_package.py (from .github/workflows/validate.yml, static-checks job). Output: OK: 10 Python files, 65 JSON files, 55 pose JSON/preview pairs, 8 README images; exit 0.

Run details
run
Arekejoker--qwen21-nsfw-pose-skill-claude-haiku-5-5-1792339015152434495
date
Oct 18, 2026
time to working
not recorded
product hosting / execution
Local command in the guest
recorded operation scope
product operation
model
claude-haiku-5-5
agent / version
claude-code / 2.1.300 (Claude Code)
platform
linux
reported cost
$0.009302
Recorded source
https://github.com/Arekejoker/qwen21-nsfw-pose-skill
Tool transport
mcp-assigned-vm
Test version
readme-03aea4c41d53955e3723c330e9927d2a9df627f6642624cc0078a97c73a53c8a
revision
eca9d7fbf8c548818f0876661b7dd489960d8f90
What the agent did

Ran the CI workflow's documented offline check: python scripts/validate_package.py (from .github/workflows/validate.yml, static-checks job). Output: OK: 10 Python files, 65 JSON files, 55 pose JSON/preview pairs, 8 README images; exit 0.

cd <host-path> && ls -la && cat README.md 2>/dev/null | head -300
cd <host-path> && git log -1 --format=%H 2>/dev/null; ls .github/workflows 2>/dev/null; which node npm python3 pip3 uv 2>&1; cat /etc/os-release | head -2
cd <host-path> && ls -R install scripts workflows | head -60 && cat install/setup.md && echo ---- && cat .github/workflows/validate.yml && echo ---- && echo $CDP_URL $<env-var> && nvidia-smi 2>&1 | head -3; free -g | head -2; df -h / | tail -1
cd <host-path> && python3 --version && python3 scripts/validate_package.py; echo "rc=$?"; echo ----; python3 scripts/doctor.py 2>&1 | head -60; echo "rc=$?"
cd <host-path> && mkdir -p .coldpass && cat > .coldpass/probe.json <<'EOF'
{
  "kind": "cli",
  "argv": ["/usr/bin/python3", "scripts/validate_package.py"],
  "cwd": "<host-path>"
}
EOF
cat > .coldpass/report.json <<'EOF'
{
  "outcome": "working",
  "operation": "Ran the CI workflow's documented offline check: python scripts/validate_package.py (from .github/workflows/validate.yml, static-checks job). Output: OK: 10 Python files, 65 JSON files, 55 pose JSON/preview pairs, 8 README images; exit 0.",
  "operation_scope": "Covers static package validation only. Image generation (README quickstart: scripts/qwen_pose_test.py) and pose transfer/PoseStudio (scripts/run_vnccs_ui.py) were NOT exercised. They need a running <email>:8188, a CUDA GPU (none in this VM), about 16GB of base models and LoRAs, the VNCCS_Utils custom node, and for route 3 a CDP browser (Python playwright package not installed; the shared CDP endpoint was not connected). scripts/doctor.py run for reference: 6 OK, 1 WARN (playwright missing), 1 FAIL (ComfyUI unreachable).",
  "steps": [
    "Read README.md and install/setup.md",
    "Read .github/workflows/validate.yml (static-checks job: python scripts/validate_package.py)",
    "python3 --version -> Python 3.12.3 (system python, no install needed)",
    "python3 scripts/validate_package.py -> OK, rc 0",
    "python3 scripts/doctor.py -> 6 OK, 1 WARN, 1 FAIL (read-only check, setup only)"
  ],
  "deviations": [
    "README quickstart 'python3 scripts/qwen_pose_test.py ... --out outputs/author_case/shower.png' not run: requires ComfyUI on 127.0.0.1:8188, a GPU, and about 16GB of models plus LoRAs. None are available in this guest (no nvidia-smi, 3GB RAM).",
    "install/setup.md sections 0-4 (model and LoRA downloads via hf-mirror.com and Civitai, VNCCS_Utils clone and pip install, pose library import into ComfyUI) not performed. ComfyUI is not installed. Multi-GB downloads were out of scope for this smoke run.",
    "install/setup.md Python playwright package install and its browser download step were not run: the task forbids downloading a browser. The shared CDP endpoint (CDP_URL) was not connected."
  ],
  "stuck_step": null,
  "main_blocker": null,
  "suggested_fix": "For full generation: provision a CUDA GPU host (12GB+ VRAM), install ComfyUI, then follow install/setup.md sections 1-4 and run the README quickstart command. Route 3 also needs the playwright package and a CDP browser pointed at CDP_URL."
}
EOF
<env-var>=<host-path> python3 -I <host-path> --validate; echo "rc=$?"