[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78AEPXM5SSF22E4E6D4BBAD":3,"skill-same:01M78AEPXM5SSF22E4E6D4BBAD":187},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":31,"commitSha":32,"skillDirectory":33,"forksCount":14,"repositoryHtmlUrl":34,"repositoryHomepageUrl":6,"ownerHtmlUrl":35,"ownerBlogUrl":36,"ownerDescription":37,"updatedAt":38,"install":39,"resources":52},"01M78AEPXM5SSF22E4E6D4BBAD","nv-generate-ct-rflow",null,"Used for generating synthetic CT volumes and masks with NV-Generate-CTMR rflow-ct. Not for production training data without review.","Solizardking\u002Ftrain2earn","Solizardking","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F108096810?v=4","01M786KDB10AYV042N0F9BD5B5","Clawd","01M786KDC5C3NPE4W44C5641EZ",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: nv-generate-ct-rflow\ndescription: Used for generating synthetic CT volumes and masks with NV-Generate-CTMR rflow-ct. Not for production training data without review.\nlicense: Apache-2.0\nallowed-tools: Bash\nmetadata:\n  author: NVIDIA MedTech Team\n  tags:\n    - MedTech\n    - CT\n    - generation\n---\n\n# NV-Generate-CT (rflow-ct)\n\n## Purpose\n- Used for generating synthetic CT volumes and masks with NV-Generate-CTMR rflow-ct. Not for production training data without review.\n- Use the wrapper exactly as documented; do not replace the upstream entrypoint with a handwritten implementation.\n- Do not write custom inference code for normal runs. The wrapper owns config staging, output paths, label mapping evidence, and validation.\n- Manifest I\u002FO: inputs are `config_infer_override`; outputs are `synthetic_ct_volumes` and `result_json`.\n\n## Instructions\n- Read `skill_manifest.yaml` before changing arguments, side effects, or validation gates.\n- Run `scripts\u002Frun_rflow_ct.py` through the documented command below; keep outputs under a caller-provided run directory.\n- If a host agent exposes `run_script`, use `run_script(\"scripts\u002Frun_rflow_ct.py\", args=[...])`; otherwise run the Bash\u002FPython command shown below.\n- Emit a single bash code block, and keep the `python -m pip install -r \"$NV_GENERATE_ROOT\u002Frequirements.txt\"` step in that same command — the runtime may be a fresh environment without `nibabel`\u002FMONAI, so dropping the install fails with `ModuleNotFoundError`.\n- Do not add `rm`, `mkdir`, or any cleanup of `--output-dir`; the wrapper creates it. Use a fresh `--output-dir` instead of deleting one.\n- Check the emitted JSON and paired verifier guidance before treating the run as evidence.\n\n## Available Scripts\n| Script | Purpose | Arguments |\n|---|---|---|\n| `scripts\u002F_anatomy.py` | Internal helper used by the primary entrypoint. | Imported only; do not call directly. |\n| `scripts\u002F_summary_card.py` | Internal helper used by the primary entrypoint. | Imported only; do not call directly. |\n| `scripts\u002Flist_anatomies.py` | Helper command for catalog or anatomy lookup. | `[--region REGION] [--filter TEXT] [--controllable]` |\n| `scripts\u002Frun_rflow_ct.py` | Primary entrypoint declared by skill_manifest.yaml. | `CONFIG_INFER.json --output-dir OUT_DIR [--random-seed N] [--version rflow-ct] [--yes]` |\n| `scripts\u002Frun_ct_mask.py` | Advanced diagnostic helper for standalone raw MAISI mask generation. | `REQUEST.json --output-dir OUT_DIR [--random-seed N] [--preflight-only] [--yes]` |\n| `scripts\u002Frun_ct_from_mask.py` | Advanced helper for CT image generation from a MAISI label mask. | `REQUEST.json --output-dir OUT_DIR [--random-seed N] [--yes]` |\n| `scripts\u002Frun_ct_image.py` | Advanced helper for CT image-only generation without paired labels. | `MODEL_CONFIG.json --output-dir OUT_DIR [--version rflow-ct] [--random-seed N] [--yes]` |\n\n## Prerequisites\n- Required environment variables: `NV_GENERATE_ROOT`.\n- Runtime requirements: GPU\u002FCUDA when declared by the manifest; Python packages listed in `runtime.side_effects.pip_packages`.\n- Side effects: writes generated outputs under the caller's `--output-dir`, may cache model assets under `~\u002F.cache\u002Fhuggingface\u002F`, and may contact `https:\u002F\u002Fhuggingface.co` or `https:\u002F\u002Fgithub.com` during setup.\n- Run commands from the repository root unless an existing section below says otherwise.\n\n## Limitations\n- This is a thin wrapper. Inference, sampling, and decoding are delegated entirely to NVIDIA-Medtech\u002FNV-Generate-CTMR's `scripts.inference`. Do not modify code under $NV_GENERATE_ROOT.\n- rflow-ct requires CUDA and ≈ 16 GB VRAM minimum for the default 256³ output_size. Larger output_size (e.g. 512×512×768) needs an A100\u002FH100.\n- Output volumes are synthetic. They are not safe to use as training data for production medtech models without an independent quality review.\n- Not for clinical deployment, clinical interpretation, autonomous diagnosis, regulatory submission.\n\n## Troubleshooting\n| Error | Cause | Fix |\n|---|---|---|\n| Missing dependency or import error | Runtime package drift from `skill_manifest.yaml`. | Install the packages declared in the manifest or use the documented setup command. |\n| Empty or schema-invalid output | Wrong input path, unsupported modality, or upstream failure. | Re-run with a known fixture and inspect the wrapper JSON plus stderr. |\n| Validation gate failure | Output violated a declared engineering invariant. | Keep the failed evidence pack and use the gate message to repair inputs or wrapper code. |\n\nWraps the upstream\n[`NVIDIA-Medtech\u002FNV-Generate-CTMR`](https:\u002F\u002Fgithub.com\u002FNVIDIA-Medtech\u002FNV-Generate-CTMR\u002Ftree\u002F61c4ec709b84cad468852243c48e250bec732074)\nrectified-flow synthesis pipeline. The wrapper does not reimplement diffusion,\nsampling, or autoencoder decoding — it shells out to the upstream\n`scripts.inference` entry point exactly as the project's README documents and\ninspects the produced image\u002Fmask pairs.\n\n## Preconditions\n\n1. If `NV_GENERATE_ROOT` already names a local checkout, the wrapper uses it\n   and records its current commit in the result. Otherwise, create the\n   recommended pinned default checkout (one-time):\n\n   ```bash\n   if [ -z \"${NV_GENERATE_ROOT:-}\" ]; then\n     export NV_GENERATE_COMMIT=61c4ec709b84cad468852243c48e250bec732074\n     export NV_GENERATE_ROOT=\"$HOME\u002F.cache\u002Fnvidia-skills\u002Fupstreams\u002FNV-Generate-CTMR-61c4ec7\"\n     if [ ! -d \"$NV_GENERATE_ROOT\u002F.git\" ]; then\n       git clone https:\u002F\u002Fgithub.com\u002FNVIDIA-Medtech\u002FNV-Generate-CTMR.git \"$NV_GENERATE_ROOT\"\n       git -C \"$NV_GENERATE_ROOT\" checkout --detach \"$NV_GENERATE_COMMIT\"\n     fi\n   fi\n   python -m pip install -r \"$NV_GENERATE_ROOT\u002Frequirements.txt\"\n   ```\n\n2. Download the `rflow-ct` weights **and** the mask-candidate datasets\n   into the clone (one-time, ≈ 5.5 GB):\n\n   ```bash\n   cd \"$NV_GENERATE_ROOT\"\n   python -m scripts.download_model_data --version rflow-ct --root_dir \".\u002F\"\n   ```\n\n   The mask candidates (`datasets\u002Fall_masks_flexible_size_and_spacing_4000`)\n   condition the diffusion sampler; omitting them via `--model_only` will\n   make the inference script fail with a missing-file error at startup.\n   The anatomy-size condition file is also part of the full CT download and is\n   needed for controllable mask generation.\n\n3. NVIDIA GPU with ≥ 16 GB VRAM and CUDA. There is no CPU fallback.\n\nFor agent-generated user run commands, prefer the short wrapper command in\nUsage. Do not prepend clone or model-download setup steps when `NV_GENERATE_ROOT`\nor the repo-local upstream cache is already present. In a fresh Python\nenvironment, still include `python -m pip install -r \"$NV_GENERATE_ROOT\u002Frequirements.txt\"`\nbefore the wrapper unless the active environment has already proven those\nimports are available; cached weights do not imply cached Python packages. Run\nthe wrapper from the medical-AI-skills repo root. If setup requires `cd \"$NV_GENERATE_ROOT\"`, return to the Medical AI Skills repo before invoking\n`skills\u002Fnv-generate-ct-rflow\u002Fscripts\u002Frun_rflow_ct.py`.\n\n## Usage\n\n```bash\nexport NV_GENERATE_ROOT=\"${NV_GENERATE_ROOT:-$HOME\u002F.cache\u002Fnvidia-skills\u002Fupstreams\u002FNV-Generate-CTMR-61c4ec7}\" && \\\npython -m pip install -r \"$NV_GENERATE_ROOT\u002Frequirements.txt\" && \\\npython skills\u002Fnv-generate-ct-rflow\u002Fscripts\u002Frun_rflow_ct.py \\\n  PATH_TO_CONFIG_INFER.json \\\n  --output-dir runs\u002Fnv_generate_ct_rflow_demo \\\n  --random-seed 0 \\\n  --version rflow-ct\n```\n\nReplace `PATH_TO_CONFIG_INFER.json` with the user's actual request\u002Fconfig\npath. Do not copy the fixture path from this document unless the user\nexplicitly asked to run that fixture. If the user says \"the case request is at\n`runs\u002F...\u002Fchest_lung_tumor_controllable.json`\", that exact path is the first\npositional argument to `scripts\u002Frun_rflow_ct.py`.\n\nThe fixture argument is a `config_infer.json` override file: it can replace\n`num_output_samples`, `body_region`, `anatomy_list`, `controllable_anatomy_size`,\n`output_size`, and `spacing`. Pass `default` to use the upstream config\nverbatim. The wrapper stages the override into the upstream tree before\nrunning.\n\n### Fixture catalog\n\n`fixtures\u002F` ships curated configs for common paired synthesis use cases: chest\nlung lobes, chest with controllable lung tumor, abdomen solid organs,\nabdomen with controllable hepatic tumor, head + cervical spine, pelvis.\nSee [`fixtures\u002FREADME.md`](fixtures\u002FREADME.md) for the full table.\n\n### Helper commands\n\n```bash\n# Browse the 132-class label_dict grouped by body region.\npython skills\u002Fnv-generate-ct-rflow\u002Fscripts\u002Flist_anatomies.py --region chest\npython skills\u002Fnv-generate-ct-rflow\u002Fscripts\u002Flist_anatomies.py --controllable\npython skills\u002Fnv-generate-ct-rflow\u002Fscripts\u002Flist_anatomies.py --filter tumor\n\n# Validate a fixture and preview cost without launching inference.\nNV_GENERATE_ROOT=$HOME\u002F.cache\u002Fnvidia-skills\u002Fupstreams\u002FNV-Generate-CTMR-61c4ec7 \\\n  python skills\u002Fnv-generate-ct-rflow\u002Fscripts\u002Frun_rflow_ct.py \\\n    skills\u002Fnv-generate-ct-rflow\u002Ffixtures\u002Fabdomen_liver_spleen.json \\\n    --output-dir runs\u002Fpreview --preflight-only\n```\n\nAdvanced helpers stay inside this skill for debugging and less-common CT\ngeneration modes. Use them only when the user explicitly asks for that mode:\n\n```bash\n# Raw MAISI mask diagnostic, useful for checking lung tumor -> label 23.\npython skills\u002Fnv-generate-ct-rflow\u002Fscripts\u002Frun_ct_mask.py \\\n  skills\u002Fnv-generate-ct-rflow\u002Ffixtures\u002Fct_mask_lung_tumor.json \\\n  --output-dir runs\u002Fct_mask_debug --preflight-only\n\n# CT image from an existing MAISI label mask with body label 200.\npython skills\u002Fnv-generate-ct-rflow\u002Fscripts\u002Frun_ct_from_mask.py \\\n  skills\u002Fnv-generate-ct-rflow\u002Ffixtures\u002Fct_from_mask_request_example.json \\\n  --output-dir runs\u002Fct_from_mask_demo\n\n# CT image-only generation without paired labels.\npython skills\u002Fnv-generate-ct-rflow\u002Fscripts\u002Frun_ct_image.py \\\n  skills\u002Fnv-generate-ct-rflow\u002Ffixtures\u002Fct_image_only_default.json \\\n  --output-dir runs\u002Fct_image_only_demo --version rflow-ct\n```\n\nThe wrapper runs preflight on every invocation (regardless of\n`--preflight-only`): config-schema bounds, anatomy names matched\nagainst the upstream label_dict, body_region in the supported set,\ncontrollable_anatomy_size constraints, upstream CT output-size\u002Fspacing\ncontracts, body-region-aware x\u002Fy FOV minimums, dataset presence under\n`$NV_GENERATE_ROOT\u002Fdatasets\u002F`, CUDA available, and an estimated peak VRAM \u002F\nwall-time. Runs estimated to exceed 5 min wall-time or 30 GB VRAM peak require\n`--yes` to proceed.\n\nEach invocation runs `python -m scripts.inference -t configs\u002Fconfig_network_rflow.json\n-i configs\u002Fconfig_infer.json -e configs\u002Fenvironment_rflow-ct.json --random-seed \u003Cs>\n--version rflow-ct`. Output evidence records the upstream git commit, model\ncheckpoint hashes, the rendered config, per-sample image\u002Fmask geometry, mask\nlabel set, image HU range summary, and per-class voxel volumes.\n\nWhen `controllable_anatomy_size` is non-empty, upstream ignores the broader\n`anatomy_list` for the saved paired label map and filters labels to the\ncontrollable anatomy names. The saved paired label values are local `1..N`\nordinals, not raw MAISI label IDs. Read `output.output_label_mapping` in\n`result_json` to map saved output labels back to source labels; for example,\noutput label `1` can represent MAISI label `23` (`lung tumor`).\nFor curated lung-tumor examples, prefer a controllable size around `0.5` or\nlarger; smaller requests such as `0.2` can produce absent or extremely small\nlabel-23 components for some seeds.\n\nFor FOV and setup details, see `references\u002Ffov-and-downloads.md`. For\nadvanced helper label-space details, see\n`references\u002Fct-mask-label-space.md` and `references\u002Fct-from-mask-format.md`.\n\n### Visual sample card\n\nAlongside the NIfTI pairs, the wrapper writes `summary.html` to the\noutput directory: a per-sample mid-slice triptych (axial \u002F coronal \u002F\nsagittal) with label overlay, plus a table of the rendered config and\nverifier-facing aggregates. Lets you eyeball the result without firing\nup 3D Slicer. Pass `--no-summary-card` to skip.\n\nAnatomy plausibility (label-set sanity, voxel HU range as CT, image\u002Fmask\ngeometry match, declared output labels present, lung-lobe HU floor) is checked by\n`verifiers\u002Fct_synthesis_quality_v1`.\n\nNot for clinical interpretation, training data for production deployment, or\nany non-synthetic-research use.\n","# NV-Generate-CT (rflow-ct)\n\n## Purpose\n- Used for generating synthetic CT volumes and masks with NV-Generate-CTMR rflow-ct. Not for production training data without review.\n- Use the wrapper exactly as documented; do not replace the upstream entrypoint with a handwritten implementation.\n- Do not write custom inference code for normal runs. The wrapper owns config staging, output paths, label mapping evidence, and validation.\n- Manifest I\u002FO: inputs are `config_infer_override`; outputs are `synthetic_ct_volumes` and `result_json`.\n\n## Instructions\n- Read `skill_manifest.yaml` before changing arguments, side effects, or validation gates.\n- Run `scripts\u002Frun_rflow_ct.py` through the documented command below; keep outputs under a caller-provided run directory.\n- If a host agent exposes `run_script`, use `run_script(\"scripts\u002Frun_rflow_ct.py\", args=[...])`; otherwise run the Bash\u002FPython command shown below.\n- Emit a single bash code block, and keep the `python -m pip install -r \"$NV_GENERATE_ROOT\u002Frequirements.txt\"` step in that same command — the runtime may be a fresh environment without `nibabel`\u002FMONAI, so dropping the install fails with `ModuleNotFoundError`.\n- Do not add `rm`, `mkdir`, or any cleanup of `--output-dir`; the wrapper creates it. Use a fresh `--output-dir` instead of deleting one.\n- Check the emitted JSON and paired verifier guidance before treating the run as evidence.\n\n## Available Scripts\n| Script | Purpose | Arguments |\n|---|---|---|\n| `scripts\u002F_anatomy.py` | Internal helper used by the primary entrypoint. | Imported only; do not call directly. |\n| `scripts\u002F_summary_card.py` | Internal helper used by the primary entrypoint. | Imported only; do not call directly. |\n| `scripts\u002Flist_anatomies.py` | Helper command for catalog or anatomy lookup. | `[--region REGION] [--filter TEXT] [--controllable]` |\n| `scripts\u002Frun_rflow_ct.py` | Primary entrypoint declared by skill_manifest.yaml. | `CONFIG_INFER.json --output-dir OUT_DIR [--random-seed N] [--version rflow-ct] [--yes]` |\n| `scripts\u002Frun_ct_mask.py` | Advanced diagnostic helper for standalone raw MAISI mask generation. | `REQUEST.json --output-dir OUT_DIR [--random-seed N] [--preflight-only] [--yes]` |\n| `scripts\u002Frun_ct_from_mask.py` | Advanced helper for CT image generation from a MAISI label mask. | `REQUEST.json --output-dir OUT_DIR [--random-seed N] [--yes]` |\n| `scripts\u002Frun_ct_image.py` | Advanced helper for CT image-only generation without paired labels. | `MODEL_CONFIG.json --output-dir OUT_DIR [--version rflow-ct] [--random-seed N] [--yes]` |\n\n## Prerequisites\n- Required environment variables: `NV_GENERATE_ROOT`.\n- Runtime requirements: GPU\u002FCUDA when declared by the manifest; Python packages listed in `runtime.side_effects.pip_packages`.\n- Side effects: writes generated outputs under the caller's `--output-dir`, may cache model assets under `~\u002F.cache\u002Fhuggingface\u002F`, and may contact `https:\u002F\u002Fhuggingface.co` or `https:\u002F\u002Fgithub.com` during setup.\n- Run commands from the repository root unless an existing section below says otherwise.\n\n## Limitations\n- This is a thin wrapper. Inference, sampling, and decoding are delegated entirely to NVIDIA-Medtech\u002FNV-Generate-CTMR's `scripts.inference`. Do not modify code under $NV_GENERATE_ROOT.\n- rflow-ct requires CUDA and ≈ 16 GB VRAM minimum for the default 256³ output_size. Larger output_size (e.g. 512×512×768) needs an A100\u002FH100.\n- Output volumes are synthetic. They are not safe to use as training data for production medtech models without an independent quality review.\n- Not for clinical deployment, clinical interpretation, autonomous diagnosis, regulatory submission.\n\n## Troubleshooting\n| Error | Cause | Fix |\n|---|---|---|\n| Missing dependency or import error | Runtime package drift from `skill_manifest.yaml`. | Install the packages declared in the manifest or use the documented setup command. |\n| Empty or schema-invalid output | Wrong input path, unsupported modality, or upstream failure. | Re-run with a known fixture and inspect the wrapper JSON plus stderr. |\n| Validation gate failure | Output violated a declared engineering invariant. | Keep the failed evidence pack and use the gate message to repair inputs or wrapper code. |\n\nWraps the upstream\n[`NVIDIA-Medtech\u002FNV-Generate-CTMR`](https:\u002F\u002Fgithub.com\u002FNVIDIA-Medtech\u002FNV-Generate-CTMR\u002Ftree\u002F61c4ec709b84cad468852243c48e250bec732074)\nrectified-flow synthesis pipeline. The wrapper does not reimplement diffusion,\nsampling, or autoencoder decoding — it shells out to the upstream\n`scripts.inference` entry point exactly as the project's README documents and\ninspects the produced image\u002Fmask pairs.\n\n## Preconditions\n\n1. If `NV_GENERATE_ROOT` already names a local checkout, the wrapper uses it\n   and records its current commit in the result. Otherwise, create the\n   recommended pinned default checkout (one-time):\n\n   ```bash\n   if [ -z \"${NV_GENERATE_ROOT:-}\" ]; then\n     export NV_GENERATE_COMMIT=61c4ec709b84cad468852243c48e250bec732074\n     export NV_GENERATE_ROOT=\"$HOME\u002F.cache\u002Fnvidia-skills\u002Fupstreams\u002FNV-Generate-CTMR-61c4ec7\"\n     if [ ! -d \"$NV_GENERATE_ROOT\u002F.git\" ]; then\n       git clone https:\u002F\u002Fgithub.com\u002FNVIDIA-Medtech\u002FNV-Generate-CTMR.git \"$NV_GENERATE_ROOT\"\n       git -C \"$NV_GENERATE_ROOT\" checkout --detach \"$NV_GENERATE_COMMIT\"\n     fi\n   fi\n   python -m pip install -r \"$NV_GENERATE_ROOT\u002Frequirements.txt\"\n   ```\n\n2. Download the `rflow-ct` weights **and** the mask-candidate datasets\n   into the clone (one-time, ≈ 5.5 GB):\n\n   ```bash\n   cd \"$NV_GENERATE_ROOT\"\n   python -m scripts.download_model_data --version rflow-ct --root_dir \".\u002F\"\n   ```\n\n   The mask candidates (`datasets\u002Fall_masks_flexible_size_and_spacing_4000`)\n   condition the diffusion sampler; omitting them via `--model_only` will\n   make the inference script fail with a missing-file error at startup.\n   The anatomy-size condition file is also part of the full CT download and is\n   needed for controllable mask generation.\n\n3. NVIDIA GPU with ≥ 16 GB VRAM and CUDA. There is no CPU fallback.\n\nFor agent-generated user run commands, prefer the short wrapper command in\nUsage. Do not prepend clone or model-download setup steps when `NV_GENERATE_ROOT`\nor the repo-local upstream cache is already present. In a fresh Python\nenvironment, still include `python -m pip install -r \"$NV_GENERATE_ROOT\u002Frequirements.txt\"`\nbefore the wrapper unless the active environment has already proven those\nimports are available; cached weights do not imply cached Python packages. Run\nthe wrapper from the medical-AI-skills repo root. If setup requires `cd \"$NV_GENERATE_ROOT\"`, return to the Medical AI Skills repo before invoking\n`skills\u002Fnv-generate-ct-rflow\u002Fscripts\u002Frun_rflow_ct.py`.\n\n## Usage\n\n```bash\nexport NV_GENERATE_ROOT=\"${NV_GENERATE_ROOT:-$HOME\u002F.cache\u002Fnvidia-skills\u002Fupstreams\u002FNV-Generate-CTMR-61c4ec7}\" && \\\npython -m pip install -r \"$NV_GENERATE_ROOT\u002Frequirements.txt\" && \\\npython skills\u002Fnv-generate-ct-rflow\u002Fscripts\u002Frun_rflow_ct.py \\\n  PATH_TO_CONFIG_INFER.json \\\n  --output-dir runs\u002Fnv_generate_ct_rflow_demo \\\n  --random-seed 0 \\\n  --version rflow-ct\n```\n\nReplace `PATH_TO_CONFIG_INFER.json` with the user's actual request\u002Fconfig\npath. Do not copy the fixture path from this document unless the user\nexplicitly asked to run that fixture. If the user says \"the case request is at\n`runs\u002F...\u002Fchest_lung_tumor_controllable.json`\", that exact path is the first\npositional argument to `scripts\u002Frun_rflow_ct.py`.\n\nThe fixture argument is a `config_infer.json` override file: it can replace\n`num_output_samples`, `body_region`, `anatomy_list`, `controllable_anatomy_size`,\n`output_size`, and `spacing`. Pass `default` to use the upstream config\nverbatim. The wrapper stages the override into the upstream tree before\nrunning.\n\n### Fixture catalog\n\n`fixtures\u002F` ships curated configs for common paired synthesis use cases: chest\nlung lobes, chest with controllable lung tumor, abdomen solid organs,\nabdomen with controllable hepatic tumor, head + cervical spine, pelvis.\nSee [`fixtures\u002FREADME.md`](fixtures\u002FREADME.md) for the full table.\n\n### Helper commands\n\n```bash\n# Browse the 132-class label_dict grouped by body region.\npython skills\u002Fnv-generate-ct-rflow\u002Fscripts\u002Flist_anatomies.py --region chest\npython skills\u002Fnv-generate-ct-rflow\u002Fscripts\u002Flist_anatomies.py --controllable\npython skills\u002Fnv-generate-ct-rflow\u002Fscripts\u002Flist_anatomies.py --filter tumor\n\n# Validate a fixture and preview cost without launching inference.\nNV_GENERATE_ROOT=$HOME\u002F.cache\u002Fnvidia-skills\u002Fupstreams\u002FNV-Generate-CTMR-61c4ec7 \\\n  python skills\u002Fnv-generate-ct-rflow\u002Fscripts\u002Frun_rflow_ct.py \\\n    skills\u002Fnv-generate-ct-rflow\u002Ffixtures\u002Fabdomen_liver_spleen.json \\\n    --output-dir runs\u002Fpreview --preflight-only\n```\n\nAdvanced helpers stay inside this skill for debugging and less-common CT\ngeneration modes. Use them only when the user explicitly asks for that mode:\n\n```bash\n# Raw MAISI mask diagnostic, useful for checking lung tumor -> label 23.\npython skills\u002Fnv-generate-ct-rflow\u002Fscripts\u002Frun_ct_mask.py \\\n  skills\u002Fnv-generate-ct-rflow\u002Ffixtures\u002Fct_mask_lung_tumor.json \\\n  --output-dir runs\u002Fct_mask_debug --preflight-only\n\n# CT image from an existing MAISI label mask with body label 200.\npython skills\u002Fnv-generate-ct-rflow\u002Fscripts\u002Frun_ct_from_mask.py \\\n  skills\u002Fnv-generate-ct-rflow\u002Ffixtures\u002Fct_from_mask_request_example.json \\\n  --output-dir runs\u002Fct_from_mask_demo\n\n# CT image-only generation without paired labels.\npython skills\u002Fnv-generate-ct-rflow\u002Fscripts\u002Frun_ct_image.py \\\n  skills\u002Fnv-generate-ct-rflow\u002Ffixtures\u002Fct_image_only_default.json \\\n  --output-dir runs\u002Fct_image_only_demo --version rflow-ct\n```\n\nThe wrapper runs preflight on every invocation (regardless of\n`--preflight-only`): config-schema bounds, anatomy names matched\nagainst the upstream label_dict, body_region in the supported set,\ncontrollable_anatomy_size constraints, upstream CT output-size\u002Fspacing\ncontracts, body-region-aware x\u002Fy FOV minimums, dataset presence under\n`$NV_GENERATE_ROOT\u002Fdatasets\u002F`, CUDA available, and an estimated peak VRAM \u002F\nwall-time. Runs estimated to exceed 5 min wall-time or 30 GB VRAM peak require\n`--yes` to proceed.\n\nEach invocation runs `python -m scripts.inference -t configs\u002Fconfig_network_rflow.json\n-i configs\u002Fconfig_infer.json -e configs\u002Fenvironment_rflow-ct.json --random-seed \u003Cs>\n--version rflow-ct`. Output evidence records the upstream git commit, model\ncheckpoint hashes, the rendered config, per-sample image\u002Fmask geometry, mask\nlabel set, image HU range summary, and per-class voxel volumes.\n\nWhen `controllable_anatomy_size` is non-empty, upstream ignores the broader\n`anatomy_list` for the saved paired label map and filters labels to the\ncontrollable anatomy names. The saved paired label values are local `1..N`\nordinals, not raw MAISI label IDs. Read `output.output_label_mapping` in\n`result_json` to map saved output labels back to source labels; for example,\noutput label `1` can represent MAISI label `23` (`lung tumor`).\nFor curated lung-tumor examples, prefer a controllable size around `0.5` or\nlarger; smaller requests such as `0.2` can produce absent or extremely small\nlabel-23 components for some seeds.\n\nFor FOV and setup details, see `references\u002Ffov-and-downloads.md`. For\nadvanced helper label-space details, see\n`references\u002Fct-mask-label-space.md` and `references\u002Fct-from-mask-format.md`.\n\n### Visual sample card\n\nAlongside the NIfTI pairs, the wrapper writes `summary.html` to the\noutput directory: a per-sample mid-slice triptych (axial \u002F coronal \u002F\nsagittal) with label overlay, plus a table of the rendered config and\nverifier-facing aggregates. Lets you eyeball the result without firing\nup 3D Slicer. Pass `--no-summary-card` to skip.\n\nAnatomy plausibility (label-set sanity, voxel HU range as CT, image\u002Fmask\ngeometry match, declared output labels present, lung-lobe HU floor) is checked by\n`verifiers\u002Fct_synthesis_quality_v1`.\n\nNot for clinical interpretation, training data for production deployment, or\nany non-synthetic-research use.\n",{"name":5,"license":15,"metadata":24,"description":7,"allowed-tools":30},{"tags":25,"author":29},[26,27,28],"MedTech","CT","generation","NVIDIA MedTech Team","Bash","main","86cc27ffd16af6cb85948f1155578e6f3f61cc4c","skillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn","https:\u002F\u002Fgithub.com\u002FSolizardking","https:\u002F\u002Fmusebook.trade","Got a mile high and never came down","2026-10-08 11:52:53",{"source":8,"skillName":5,"commands":40,"prompt":50,"zipAvailable":51},[41,44,47],{"tool":42,"command":43},"npx","npx skills add Solizardking\u002Ftrain2earn --skill nv-generate-ct-rflow",{"tool":45,"command":46},"bunx","bunx skills add Solizardking\u002Ftrain2earn --skill nv-generate-ct-rflow",{"tool":48,"command":49},"pnpm","pnpm dlx skills add Solizardking\u002Ftrain2earn --skill nv-generate-ct-rflow","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002FSKILL.md",true,[53,59,63,67,71,75,79,83,87,91,95,99,103,107,111,115,119,125,129,133,137,141,145,149,153,157,161,167,171,175,179,183],{"path":54,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":57,"sourceHtmlUrl":58},"BENCHMARK.md","REFERENCE","参考文档",4029,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002FBENCHMARK.md",{"path":60,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":61,"sourceHtmlUrl":62},"evals\u002Fevals.json",1386,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Fevals\u002Fevals.json",{"path":64,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":65,"sourceHtmlUrl":66},"fixtures\u002Fabdomen_hepatic_tumor.json",787,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Ffixtures\u002Fabdomen_hepatic_tumor.json",{"path":68,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":69,"sourceHtmlUrl":70},"fixtures\u002Fabdomen_liver_spleen.json",747,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Ffixtures\u002Fabdomen_liver_spleen.json",{"path":72,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":73,"sourceHtmlUrl":74},"fixtures\u002Fchest_lung_lobes.json",680,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Ffixtures\u002Fchest_lung_lobes.json",{"path":76,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":77,"sourceHtmlUrl":78},"fixtures\u002Fchest_lung_tumor_controllable.json",850,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Ffixtures\u002Fchest_lung_tumor_controllable.json",{"path":80,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":81,"sourceHtmlUrl":82},"fixtures\u002Fct_from_mask_request_example.json",204,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Ffixtures\u002Fct_from_mask_request_example.json",{"path":84,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":85,"sourceHtmlUrl":86},"fixtures\u002Fct_image_only_default.json",284,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Ffixtures\u002Fct_image_only_default.json",{"path":88,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":89,"sourceHtmlUrl":90},"fixtures\u002Fct_mask_lung_tumor.json",413,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Ffixtures\u002Fct_mask_lung_tumor.json",{"path":92,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":93,"sourceHtmlUrl":94},"fixtures\u002Fdefault_config_infer.json",679,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Ffixtures\u002Fdefault_config_infer.json",{"path":96,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":97,"sourceHtmlUrl":98},"fixtures\u002Fhead_brain.json",710,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Ffixtures\u002Fhead_brain.json",{"path":100,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":101,"sourceHtmlUrl":102},"fixtures\u002Fpelvis.json",707,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Ffixtures\u002Fpelvis.json",{"path":104,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":105,"sourceHtmlUrl":106},"fixtures\u002FREADME.md",4458,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Ffixtures\u002FREADME.md",{"path":108,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":109,"sourceHtmlUrl":110},"references\u002Fct-from-mask-format.md",672,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Freferences\u002Fct-from-mask-format.md",{"path":112,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":113,"sourceHtmlUrl":114},"references\u002Fct-mask-label-space.md",761,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Freferences\u002Fct-mask-label-space.md",{"path":116,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":117,"sourceHtmlUrl":118},"references\u002Ffov-and-downloads.md",2077,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Freferences\u002Ffov-and-downloads.md",{"path":120,"resourceType":121,"resourceTypeDesc":122,"sizeBytes":123,"sourceHtmlUrl":124},"scripts\u002F_anatomy.py","SCRIPT","脚本",12083,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Fscripts\u002F_anatomy.py",{"path":126,"resourceType":121,"resourceTypeDesc":122,"sizeBytes":127,"sourceHtmlUrl":128},"scripts\u002F_summary_card.py",11115,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Fscripts\u002F_summary_card.py",{"path":130,"resourceType":121,"resourceTypeDesc":122,"sizeBytes":131,"sourceHtmlUrl":132},"scripts\u002Flist_anatomies.py",4590,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Fscripts\u002Flist_anatomies.py",{"path":134,"resourceType":121,"resourceTypeDesc":122,"sizeBytes":135,"sourceHtmlUrl":136},"scripts\u002Frun_ct_from_mask.py",17916,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Fscripts\u002Frun_ct_from_mask.py",{"path":138,"resourceType":121,"resourceTypeDesc":122,"sizeBytes":139,"sourceHtmlUrl":140},"scripts\u002Frun_ct_image.py",20906,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Fscripts\u002Frun_ct_image.py",{"path":142,"resourceType":121,"resourceTypeDesc":122,"sizeBytes":143,"sourceHtmlUrl":144},"scripts\u002Frun_ct_mask.py",20445,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Fscripts\u002Frun_ct_mask.py",{"path":146,"resourceType":121,"resourceTypeDesc":122,"sizeBytes":147,"sourceHtmlUrl":148},"scripts\u002Frun_rflow_ct.py",39542,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Fscripts\u002Frun_rflow_ct.py",{"path":150,"resourceType":121,"resourceTypeDesc":122,"sizeBytes":151,"sourceHtmlUrl":152},"scripts\u002Fwrapper_utils.py",2114,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Fscripts\u002Fwrapper_utils.py",{"path":154,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":155,"sourceHtmlUrl":156},"skill_manifest.yaml",8088,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Fskill_manifest.yaml",{"path":158,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":159,"sourceHtmlUrl":160},"skill-card.md",4273,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Fskill-card.md",{"path":162,"resourceType":163,"resourceTypeDesc":164,"sizeBytes":165,"sourceHtmlUrl":166},"skill.oms.sig","OTHER","其他",11353,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Fskill.oms.sig",{"path":168,"resourceType":121,"resourceTypeDesc":122,"sizeBytes":169,"sourceHtmlUrl":170},"tests\u002Ftest_run_ct_from_mask.py",2760,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Ftests\u002Ftest_run_ct_from_mask.py",{"path":172,"resourceType":121,"resourceTypeDesc":122,"sizeBytes":173,"sourceHtmlUrl":174},"tests\u002Ftest_run_ct_image.py",3261,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Ftests\u002Ftest_run_ct_image.py",{"path":176,"resourceType":121,"resourceTypeDesc":122,"sizeBytes":177,"sourceHtmlUrl":178},"tests\u002Ftest_run_ct_mask.py",2945,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Ftests\u002Ftest_run_ct_mask.py",{"path":180,"resourceType":121,"resourceTypeDesc":122,"sizeBytes":181,"sourceHtmlUrl":182},"tests\u002Ftest_run_rflow_ct.py",12282,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Ftests\u002Ftest_run_rflow_ct.py",{"path":184,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":185,"sourceHtmlUrl":186},"validators\u002Foutput_schema.json",6571,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fnvidia\u002Fnv-generate-ct-rflow\u002Fvalidators\u002Foutput_schema.json",[188,196,204,212,219,227],{"skillNo":189,"name":190,"nameZh":6,"description":191,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":192,"licenseFileUrl":6,"sourceHtmlUrl":193,"categories":194,"keywords":195,"publishedAt":19},"01M78B94RSCAKMH1QS79VE6DG9","testing","For testing with Light Protocol programs and clients on localnet, devnet, and mainnet validation.","MIT","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Ftesting\u002FSKILL.md",[],[],{"skillNo":197,"name":198,"nameZh":6,"description":199,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":192,"licenseFileUrl":200,"sourceHtmlUrl":201,"categories":202,"keywords":203,"publishedAt":19},"01M78B81BKY8RX12HETPG8YKWK","qedgen","Formally verify programs by writing Lean 4 proofs. Trigger this skill whenever the user wants to formally verify code, generate Lean 4 proofs, prove properties about algorithms or smart contracts, verify invariants, convert program logic into formal specifications, or anything involving Lean 4 and formal verification. Also trigger when the user mentions \"qedgen\", \"lean proof\", \"formal proof\", \"verify my code\", \"prove correctness\", \"formal verification\", or wants mathematical guarantees about their implementation.","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fsolana-formal-verification\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fsolana-formal-verification\u002FSKILL.md",[],[],{"skillNo":205,"name":206,"nameZh":6,"description":207,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":192,"licenseFileUrl":208,"sourceHtmlUrl":209,"categories":210,"keywords":211,"publishedAt":19},"01M78B7Q7C5K54E01DN83A60KG","solana-dev","Use when user asks to \"build a Solana dapp\", \"write an Anchor program\", \"create a token\", \"debug Solana errors\", \"set up wallet connection\", \"test my Solana program\", \"deploy to devnet\", or \"explain Solana concepts\" (rent, accounts, PDAs, CPIs, etc.). Also use for quick on-chain lookups via public RPC + curl — \"what's the balance of \u003Cwallet>\", \"look up transaction \u003Csig>\", \"token balance for \u003Caccount>\", \"check this address on mainnet\u002Fdevnet\". End-to-end Solana development playbook covering wallet connection, Anchor\u002FPinocchio programs, Codama client generation, LiteSVM\u002FMollusk\u002FSurfpool testing, security checklists, and JSON-RPC curl lookups against public clusters. Integrates with the Solana MCP server for live documentation search. Prefers framework-kit (@solana\u002Fclient + @solana\u002Freact-hooks) for UI, wallet-standard-first connection (incl. ConnectorKit), @solana\u002Fkit for client\u002FRPC code, and @solana\u002Fweb3-compat for legacy boundaries.","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fsolana-dev\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fsolana-dev\u002FSKILL.md",[],[],{"skillNo":213,"name":214,"nameZh":6,"description":215,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":192,"licenseFileUrl":6,"sourceHtmlUrl":216,"categories":217,"keywords":218,"publishedAt":19},"01M78B7HXD6N6GWKSVRQ8F6Q31","solana-common-errors","Diagnose and fix common Solana development errors — GLIBC mismatches, Anchor version conflicts, cargo build-sbf \u002F platform-tools failures, LiteSVM issues, edition2024 crate pins, RPC airdrops, and Anchor 0.29→0.32 migrations. Use when builds fail, Anchor CLI won't install, or CI breaks on Debian\u002FUbuntu.","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fsolana-common-errors\u002FSKILL.md",[],[],{"skillNo":220,"name":221,"nameZh":6,"description":222,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":223,"sourceHtmlUrl":224,"categories":225,"keywords":226,"publishedAt":19},"01M78B6T7YSYWQ0N4DP40QDXEZ","skill-creator","Create or update AgentSkills. Use when designing, structuring, or packaging skills with scripts, references, and assets.","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fskill-creator\u002Flicense.txt","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fskill-creator\u002FSKILL.md",[],[],{"skillNo":228,"name":229,"nameZh":6,"description":230,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":192,"licenseFileUrl":6,"sourceHtmlUrl":231,"categories":232,"keywords":233,"publishedAt":19},"01M78B1VFT9X5AND7AE1HZPZGF","phantom-wallet-mcp","Execute wallet operations through the Phantom MCP server. Use when the user wants to interact with their Phantom wallet directly — get addresses, transfer SOL or SPL tokens, buy\u002Fswap tokens, sign transactions, and sign messages across Solana, Ethereum, Bitcoin, and Sui. Requires the @phantom\u002Fmcp-server to be configured as an MCP server.","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Ftrain2earn\u002Fblob\u002Fmain\u002Fskillhub\u002Fskills\u002Fphantom-wallet-mcp\u002FSKILL.md",[],[]]