[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78HYXJRRCFMSF4WH52ZYBWT":3,"skill-same:01M78HYXJRRCFMSF4WH52ZYBWT":72},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":31,"commitSha":32,"skillDirectory":33,"forksCount":13,"repositoryHtmlUrl":34,"repositoryHomepageUrl":35,"ownerHtmlUrl":36,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":37,"install":38,"resources":51},"01M78HYXJRRCFMSF4WH52ZYBWT","jetson-customize-nvpmodel",null,"Use when you need to add, remove, edit, list, or change the boot default of an nvpmodel power mode on a Jetson\u002FTegra (Orin, Thor) target. Triggers: edit power mode, tune frequency caps.","matheussluzz\u002FZeuZ-Agent","matheussluzz","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F209255758?v=4","01M78HKHH1D1ASC9KC9BG583DH","01M78HKHHP6W7XRXQM2SBCAM3R",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fmatheussluzz\u002FZeuZ-Agent\u002Fblob\u002Fmain\u002Fcatalog\u002Fbundles\u002Fnvidia\u002Fjetson-customize-nvpmodel\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: jetson-customize-nvpmodel\ndescription: >-\n  Use when you need to add, remove, edit, list, or change the boot\n  default of an nvpmodel power mode on a Jetson\u002FTegra (Orin, Thor)\n  target. Triggers: edit power mode, tune frequency caps.\nversion: 0.0.1\nlicense: \"Apache-2.0\"\nmetadata:\n  data-classification: public\n  author: \"Jetson Team\"\n  tags:\n    - power\n    - nvpmodel\n    - profile\n  domain: power\n---\n\n# Modify nvpmodel Power Mode (BSP-side)\n\n## Purpose\n\nEdit the per-board nvpmodel configuration so the device boots with\nthe desired power-mode set, CPU\u002FGPU\u002FEMC\u002FTPC clamps, and default\nmode. BSP-side only — all writes land in the overlay tracker, the\nupstream `bsp_image` copy is read-only.\n\nThis skill handles BSP-side edits to the per-board nvpmodel configuration file: adding modes, removing modes, editing CPU\u002FGPU\u002FEMC\u002FTPC clamps, and changing the boot default. Applies on Jetson \u002F Tegra platforms (T234 Orin, T264 Thor).\n\n## File format (canonical, per the BSP file header)\n\n```\n# 1. PARAM definitions — declare named knobs and their sysfs paths\n\u003C PARAM TYPE=FILE  NAME=\u003Cparam_name> >\n\u003Carg_name> \u003C\u002Fabsolute\u002Fsysfs\u002Fpath>\n...\n\u003C PARAM TYPE=CLOCK NAME=\u003Cparam_name> >\nFREQ_TABLE        \u003C\u002Fsysfs\u002F...\u002Favailable_frequencies>\nMAX_FREQ          \u003C\u002Fsysfs\u002F...\u002Fmax_freq>\nMIN_FREQ          \u003C\u002Fsysfs\u002F...\u002Fmin_freq>\nFREQ_TABLE_KNEXT  \u003C\u002Fsysfs\u002F...\u002Favailable_frequencies>     # kernel-NEXT variant\nMAX_FREQ_KNEXT    \u003C\u002Fsysfs\u002F...\u002Fmax_freq>\nMIN_FREQ_KNEXT    \u003C\u002Fsysfs\u002F...\u002Fmin_freq>\n\n# 2. POWER_MODEL definitions — one block per profile\n\u003C POWER_MODEL ID=\u003Cint> NAME=\u003Cstring> >\nPARAM_NAME ARG_NAME \u003Cvalue>\n...\n\n# 3. PM_CONFIG — mandatory; selects boot default\n\u003C PM_CONFIG DEFAULT=\u003Cid> >\n```\n\nRules:\n\n- For `TYPE=FILE`, `\u003Cvalue>` is a **string** (`0`, `1`, `on`, `auto`, …).\n- For `TYPE=CLOCK`, `\u003Cvalue>` is an **integer** (Hz for clocks, raw integer for masks).\n- `-1` for a CLOCK value means **INT_MAX** (no cap).\n- The header `\u003C … >` line must start at column 0 with a space after `\u003C` and before `>`. Strict.\n- Every `PARAM_NAME` referenced in a POWER_MODEL must already be declared above.\n- **Frequency values must come from the kernel's `available_frequencies`** table for that clock — values not in the table get silently rounded.\n- `CORE_0` cannot be offlined; at least one CPU core must remain online in every profile.\n- **Copy PARAM names verbatim** from an existing POWER_MODEL block in the per-board file — chip families differ (T234 Orin uses `CPU_A78_\u003Cn>` for CPU clusters; T264 Thor uses a different convention). Don't invent.\n\n## Prerequisites\n\nResolve the active profile per\n[`..\u002F..\u002Fcontext\u002Ftarget-platform-contract.md`](..\u002F..\u002Fcontext\u002Ftarget-platform-contract.md).\nRefuse and route in these cases:\n\n| Condition | Refuse with |\n|---|---|\n| No active profile, or `active: NA` | Route to `\u002Fjetson-set-target` or `\u002Fjetson-init-target`. |\n| Profile lacks `bsp_image:` block | Route to `\u002Fjetson-init-image`. |\n| `\u003Cbsp_image.root_path>\u002FLinux_for_Tegra\u002F` missing | Route to `\u002Fjetson-init-image`. |\n| `\u003Csource.root_path>\u002FLinux_for_Tegra\u002F` missing or not a git repo | Route to `\u002Fjetson-init-source`. |\n\nResolve paths:\n\n- `\u003Cbsp_image.root_path>` from `bsp_image.root_path:` if present, else `\u003Cworkspace>\u002FImage`.\n- `\u003Csource.root_path>` from `source.root_path:` if present, else `\u003Cworkspace>\u002FSource`.\n\n`\u003Cbsp_image.root_path>` is **read-only** for this skill; every write lands\nunder `\u003Csource.root_path>` (the overlay tracker). This is the workflow\ninvariant in\n[`..\u002F..\u002Fcontext\u002Fbsp-customization-workflow.md#workflow-invariants`](..\u002F..\u002Fcontext\u002Fbsp-customization-workflow.md#workflow-invariants) —\nhand-editing upstream silently destroys the diff trail and makes\n`\u002Fjetson-promote-image` a noop.\n\n## The per-board file\n\nThe conf this skill edits has the relative path:\n\n```\nLinux_for_Tegra\u002Frootfs\u002Fetc\u002Fnvpmodel\u002Fnvpmodel_\u003Cactive-sku>.conf\n```\n\nIt lives in **two** roots; the skill walks both:\n\n| Role | Location | Skill writes? |\n|---|---|---|\n| Detection + pristine source | `\u003Cbsp_image.root_path>\u002FLinux_for_Tegra\u002Frootfs\u002Fetc\u002Fnvpmodel\u002F` | no — read-only |\n| Overlay edit target + git commit | `\u003Csource.root_path>\u002FLinux_for_Tegra\u002Frootfs\u002Fetc\u002Fnvpmodel\u002F` | yes |\n\nSubsequent sections refer to **the per-board file** to mean the overlay copy\nunder `\u003Csource.root_path>`. Operations 1–4 all read, edit, and save against\nthat overlay copy. The `\u003Cbsp_image.root_path>` copy is read once during the\n\"Resolving `\u003Cactive-sku>`\" detection step and once during the\n[Overlay edit recipe](#overlay-edit-recipe-apply-before-any-operation)'s\npristine-import step, then never touched again.\n\n### Resolving `\u003Cactive-sku>` — which file to edit\n\nThe filename is **not** always `module.id + \"_\" + module.sku`. Variants exist:\n\n- `nvpmodel_p3767_0000_super.conf` (super-mode SKU variant)\n- `nvpmodel_igx_orin.conf`, `nvpmodel_igx_orin_safety.conf` (IGX, no SKU number)\n\nAt boot, `nvpower.sh` (at `Linux_for_Tegra\u002Frootfs\u002Fetc\u002Fsystemd\u002Fnvpower.sh`) reads the kernel DTB's root `compatible` string and maps it — plus super\u002Fsafety state — to the nvpmodel filename. Replicate that mapping **BSP-side**, against `\u003Cbsp_image.root_path>`:\n\n1. Resolve the SKU-correct kernel DTB under `\u003Cbsp_image.root_path>\u002FLinux_for_Tegra\u002F` and read its root `compatible` (detect kernel DTB from the active flash conf).\n2. Read `\u003Cbsp_image.root_path>\u002FLinux_for_Tegra\u002Frootfs\u002Fetc\u002Fsystemd\u002Fnvpower.sh` to find the compatible-string → conf mapping, and apply it to the compatible from the previous step (factoring in super\u002Fsafety flags).\n3. Verify the resolved `nvpmodel_\u003C...>.conf` exists under `\u003Cbsp_image.root_path>\u002FLinux_for_Tegra\u002Frootfs\u002Fetc\u002Fnvpmodel\u002F`.\n\nShortcut: filter `\u003Cbsp_image.root_path>\u002FLinux_for_Tegra\u002Frootfs\u002Fetc\u002Fnvpmodel\u002F` to `nvpmodel_\u003Cmodule.id>_\u003Cmodule.sku>*.conf`; for super-mode flash configs pick the `_super` variant. Use only when unambiguous; otherwise fall back to the DTB method.\n\nDon't blindly compose `nvpmodel_\u003Cid>_\u003Csku>.conf` — verify the file actually exists.\n\n### Propagation set — confs to keep in sync\n\nThe active-SKU file is rarely the only conf that should carry a customization. After editing it, **apply the same edit to every sibling in the propagation set** so the change survives regardless of which module \u002F baseboard SKU is booted:\n\n- **The reference platform's nvpmodel conf** — the upstream conf the active file was forked from (resolve via `reference_devkit` in the active profile, or `jetson-derive-carrier` fork ancestry). For a BSP that contains only the reference (no derived carriers), this is the same file as the active and the rule reduces to a no-op.\n- **Every carrier-derived nvpmodel conf** — each `nvpmodel_*.conf` produced by `jetson-derive-carrier` for a custom carrier on top of the same module SKU.\n- **`_super` siblings** when present (e.g. `nvpmodel_p3767_0000_super.conf`) — apply the structural edit (new POWER_MODEL block, removed block, `PM_CONFIG DEFAULT=` flip, NAME rename) **but preserve the super conf's higher MAX_FREQ \u002F MIN_FREQ caps**. Those elevated caps are the whole reason the `_super` variant exists; a blanket content overwrite from the non-super file would silently flatten the super envelope.\n\n**\"Apply the same edit\" ≠ blanket file copy.** Port the changed POWER_MODEL \u002F PARAM lines into each sibling; preserve every other line. Blanket-copying is safe **only** when both confs were byte-identical before the edit. Otherwise re-validate per the Rules on each target (available-frequencies table for clock values, ID uniqueness, `PM_CONFIG DEFAULT=` references a present ID) — sibling confs may carry SKU-specific `available_frequencies` tables that don't accept the active file's frequency values.\n\n## Overlay edit recipe (apply before any Operation)\n\nFollow the canonical\n[Off-skill edits recipe](..\u002F..\u002Fcontext\u002Fbsp-customization-workflow.md#off-skill-edits)\nin the workflow doc — pristine import + customization commit pair, both\ngated by the preview gate. Apply once per run, covering every per-board\nfile the run touches (the active conf plus every sibling in the\n[Propagation set](#propagation-set--confs-to-keep-in-sync)).\n\nConcrete substitutions for this skill:\n\n- `\u003Crel>\u002F\u003Cfile>` is `rootfs\u002Fetc\u002Fnvpmodel\u002F\u003Cconf>`.\n- Suggested pristine-import message:\n  `import pristine: \u003Ccomma-separated rel paths of imported confs>`,\n  body `Source: \u003Cbsp_image.root_path>\u002FLinux_for_Tegra\u002F (BSP \u003Cbsp_image.version>)`.\n- Suggested customization-commit header:\n  `jetson-customize-nvpmodel: \u003Csummary>`,\n  body lines like `nvpmodel_p3767_0001.conf: PM_CONFIG DEFAULT 2 -> 0 (MAXN)`.\n\n## Instructions\n\nPick the operation that matches the user's intent and follow the\nmatching subsection. All write-side operations (1–4) must first\napply the [Overlay edit recipe](#overlay-edit-recipe-apply-before-any-operation).\n\n- Operation 1 — Add a new power mode (drives the NVIDIA Power Estimator).\n- Operation 2 — Remove a power mode.\n- Operation 3 — Edit an existing power mode (clamps, core-online, NAME).\n- Operation 4 — Change the boot default (`PM_CONFIG DEFAULT=`).\n- Operation 5 — List defined power modes (read-only).\n\nAfter any write-side operation, run the Deploy chain (`## Deploy`)\nto land the change on the device.\n\n## Examples\n\nAdd a new 30 W power mode from a Power Estimator export and pin it\nas the boot default on a P3767-0001 target:\n\n```\n\u002Fjetson-customize-nvpmodel\n> add a new POWER_MODEL from ~\u002FDownloads\u002Fnvpmodel_30W.conf as ID 8 NAME \"30W_CUSTOM\"\n> set PM_CONFIG DEFAULT to 8\n```\n\nCap a Thor target to the MAXN power envelope by flipping the boot\ndefault (no envelope tuning needed):\n\n```\n\u002Fjetson-customize-nvpmodel\n> set PM_CONFIG DEFAULT to the MAXN profile's ID\n```\n\nList which power modes the active SKU currently defines and which\nis the boot default:\n\n```\n\u002Fjetson-customize-nvpmodel\n> list defined power modes\n```\n\n## Operation 1 — Add a new power mode\n\nApply the [Overlay edit recipe](#overlay-edit-recipe-apply-before-any-operation) first.\n\n### Generate the profile via the Power Estimator\n\n**Must not interpolate or extrapolate frequencies from existing profiles to estimate power.**\n\nFor any non-stock power envelope (custom budget, custom workload), use the **NVIDIA Power Estimator**: `https:\u002F\u002Fjetson-tools.nvidia.com\u002Fpowerestimator\u002F`\n\n- Pick the exact module SKU and JetPack release.\n- Enter workload (CPU cores, GPU usage, EMC, codecs, camera, display).\n- Estimate power budget.\n- Download custom `nvpmodel.conf`.\n- Share the downloaded `nvpmodel.conf` file path.\n\n### Append the POWER_MODEL block\n\nIn the per-board file, insert the new block **after** the last `\u003C PARAM … >` declaration (POWER_MODEL must reference declared PARAMs) and **before** the final `\u003C PM_CONFIG ... >` line. The block structure is the **File format** shown above; frequency values come from the Power Estimator output (see \"Generate the profile via the Power Estimator\").\n\nCheck `ID` is unique; `NAME` should be uppercase, no whitespace. Each numeric `MAX_FREQ` \u002F `MIN_FREQ` value must satisfy the `available_frequencies` rule (see Rules). If you intend this mode to be the boot default, follow Operation 4.\n\n## Operation 2 — Remove a power mode\n\nApply the [Overlay edit recipe](#overlay-edit-recipe-apply-before-any-operation) first.\n\n1. In the per-board file, delete the entire `\u003C POWER_MODEL ID=\u003Cn> NAME=... >` block including all its parameter lines, up to (but not including) the next `\u003C POWER_MODEL ... >` or `\u003C PM_CONFIG ... >` marker.\n2. If the deleted ID matches the current `DEFAULT=\u003Cid>` value in the trailing `\u003C PM_CONFIG … >` line, point `DEFAULT=` at a remaining ID — otherwise nvpmodel will fail to apply a default at boot.\n3. Search for hard-coded references in the rootfs scripts before declaring the change safe:\n   ```bash\n   grep -rn \"nvpmodel -m\" \\\n     Linux_for_Tegra\u002Frootfs\u002Fetc \\\n     Linux_for_Tegra\u002Frootfs\u002Fopt 2>\u002Fdev\u002Fnull\n   ```\n\nID gaps are legal — you don't have to renumber remaining modes.\n\n## Operation 3 — Edit an existing power mode\n\nApply the [Overlay edit recipe](#overlay-edit-recipe-apply-before-any-operation) first.\n\n1. In the per-board file, edit the parameter lines inside the block. Keep `\u003Cparam>` and `\u003Carg>` names exactly matching the PARAM declarations.\n2. **If the edit changes the power envelope** (any CPU\u002FGPU\u002FEMC\u002FPVA\u002FDLA `MAX_FREQ` \u002F `MIN_FREQ`, core-online count for a freq-bound mode, or TPC mask), re-run the **NVIDIA Power Estimator** (see \"Generate the profile via the Power Estimator\") to ground the new frequencies in a real per-component model. Do not interpolate from neighboring modes.\n3. Validate clock values per the `available_frequencies` rule (see Rules).\n\nEdits that only toggle core-online flags within an already-validated envelope, or only change `NAME=`, don't need the Power Estimator pass.\n\n## Operation 4 — Change the boot default\n\nApply the [Overlay edit recipe](#overlay-edit-recipe-apply-before-any-operation) first.\n\nEdit the `\u003C PM_CONFIG DEFAULT=\u003Cid> >` line at the bottom of the per-board file.\n\n`\u003Cid>` must reference an existing `\u003C POWER_MODEL ID=\u003Cid> … >` block in the same file. Pointing at an undefined ID makes nvpmodel fail to apply at boot.\n\n## Operation 5 — List defined power modes\n\nThis is a read-only operation; no overlay-tracker setup is needed. Run\nagainst whichever copy you want to inspect (`\u003Cbsp_image.root_path>\u002F...`\nfor the pristine state, `\u003Csource.root_path>\u002F...` for the post-edit state):\n\n```bash\ngrep -E '^\u003C POWER_MODEL ' \u003Cper-board file>\ngrep -E '^\u003C PM_CONFIG '   \u003Cper-board file>\n```\n\nThe first command prints every `ID`\u002F`NAME`; the second prints the current boot default. On a running target, `nvpmodel -p --verbose` (or `nvpmodel -q`) is authoritative.\n\n## Limitations\n\n- BSP-side scope only — this skill never invokes `nvpmodel -m` on a\n  running target. Live mode switching requires reboot via Deploy,\n  or the side-channel `scp + nvpmodel -m` flow described in `## Deploy`.\n- Edits land in the overlay copy under `\u003Csource.root_path>` only;\n  the `\u003Cbsp_image.root_path>` copy is read-only and is rewritten by\n  `\u002Fjetson-promote-image`. Hand-editing `bsp_image` is silently lost\n  on the next `\u002Fjetson-init-image` re-extract.\n- Frequency values must come from the kernel's `available_frequencies`\n  table for the relevant clock — values outside the table get\n  silently rounded. This skill does not validate against a live\n  target's table; trust the per-board file's existing values and\n  Power Estimator output.\n- Non-trivial envelope edits (any CPU\u002FGPU\u002FEMC\u002FPVA\u002FDLA `MAX_FREQ` \u002F\n  `MIN_FREQ` change) require the NVIDIA Power Estimator\n  (`https:\u002F\u002Fjetson-tools.nvidia.com\u002Fpowerestimator\u002F`) — interpolation\n  or extrapolation between existing modes is not supported.\n- Propagation across siblings is partial by design: only the changed\n  `POWER_MODEL` \u002F `PARAM` lines are ported, never a blanket file\n  overwrite, since `_super` siblings carry elevated `MAX_FREQ`\u002F`MIN_FREQ`\n  caps that must stay intact.\n- PARAM names are chip-family-specific (e.g. `CPU_A78_\u003Cn>` on T234,\n  different on T264). Copy verbatim from existing POWER_MODEL blocks;\n  invented names silently fail to apply.\n\n## Troubleshooting\n\n| Error | Cause | Solution |\n|---|---|---|\n| nvpmodel fails to apply default at boot | `PM_CONFIG DEFAULT=\u003Cid>` references a deleted or missing POWER_MODEL ID | Point `DEFAULT=` at an existing `\u003C POWER_MODEL ID=\u003Cn> ... >` in the same file. |\n| Frequency value silently doesn't take effect | Value not in the kernel's `available_frequencies` table for that clock | Replace with the nearest legal value from the running target's `available_frequencies` (or Power Estimator output). |\n| `MAX_FREQ -1` does nothing | `-1` on a `TYPE=FILE` PARAM (only valid for `TYPE=CLOCK`) | Use `-1` only on CLOCK params; for FILE params, write the actual string the sysfs node expects. |\n| Module dies \u002F fails to boot after edit | `CORE_0` was offlined, or the mode dropped below the platform's minimum quiescent envelope | Keep `CORE_0` online; floor `MIN_FREQ` per the Power Estimator's lowest-power profile for the SKU. |\n| `nvpmodel -m \u003Cid>` returns \"ID not found\" at runtime | Mode ID was removed or renumbered but a rootfs script still references the old ID | `grep -rn 'nvpmodel -m' rootfs\u002Fetc rootfs\u002Fopt` and update \u002F remove the stale references. |\n| Change vanished after `\u002Fjetson-init-image` re-extract | Edit landed in `\u003Cbsp_image.root_path>` instead of `\u003Csource.root_path>` overlay | Re-apply via the [Overlay edit recipe](#overlay-edit-recipe-apply-before-any-operation) so the change is committed in the overlay tracker. |\n| `_super` sibling lost elevated caps after propagation | Blanket file copy clobbered `_super`'s higher `MAX_FREQ`\u002F`MIN_FREQ` | Port only the structural change; preserve `_super`'s caps per [Propagation set](#propagation-set--confs-to-keep-in-sync). |\n| Parser rejects new POWER_MODEL block | Header `\u003C … >` not at column 0, missing space after `\u003C` or before `>`, or PARAM not declared earlier | Restore strict header formatting; ensure every referenced `PARAM_NAME` appears in a `\u003C PARAM ... >` block above. |\n\n## Deploy\n\nThe customization commit in the overlay tracker does not reach the device\non its own. The Deploy chain:\n\n1. **`\u002Fjetson-promote-image`** — copies every tracked file in the overlay\n   into `\u003Cbsp_image.root_path>\u002FLinux_for_Tegra\u002F`. Diff-aware (skip\n   byte-identical); uses `sudo cp -p` for `rootfs\u002F*` destinations.\n2. **`\u002Fjetson-flash-image`** — flashes the updated `bsp_image` to the\n   device.\n3. (Alternate, no flash) Copy `\u003Csource.root_path>\u002FLinux_for_Tegra\u002Frootfs\u002Fetc\u002Fnvpmodel\u002F\u003Cconf>`\n   directly to the running target's `\u002Fetc\u002Fnvpmodel\u002F\u003Cconf>`, then\n   `sudo nvpmodel -m \u003Cid>` (or reboot to pick up the new `DEFAULT=`).\n\nEditing `\u003Csource.root_path>\u002F...` without committing — or editing\n`\u003Cbsp_image.root_path>\u002F...` directly — does nothing for `\u002Fjetson-promote-image`\nand is silently lost on the next `\u002Fjetson-init-image` re-extract.\n","# Modify nvpmodel Power Mode (BSP-side)\n\n## Purpose\n\nEdit the per-board nvpmodel configuration so the device boots with\nthe desired power-mode set, CPU\u002FGPU\u002FEMC\u002FTPC clamps, and default\nmode. BSP-side only — all writes land in the overlay tracker, the\nupstream `bsp_image` copy is read-only.\n\nThis skill handles BSP-side edits to the per-board nvpmodel configuration file: adding modes, removing modes, editing CPU\u002FGPU\u002FEMC\u002FTPC clamps, and changing the boot default. Applies on Jetson \u002F Tegra platforms (T234 Orin, T264 Thor).\n\n## File format (canonical, per the BSP file header)\n\n```\n# 1. PARAM definitions — declare named knobs and their sysfs paths\n\u003C PARAM TYPE=FILE  NAME=\u003Cparam_name> >\n\u003Carg_name> \u003C\u002Fabsolute\u002Fsysfs\u002Fpath>\n...\n\u003C PARAM TYPE=CLOCK NAME=\u003Cparam_name> >\nFREQ_TABLE        \u003C\u002Fsysfs\u002F...\u002Favailable_frequencies>\nMAX_FREQ          \u003C\u002Fsysfs\u002F...\u002Fmax_freq>\nMIN_FREQ          \u003C\u002Fsysfs\u002F...\u002Fmin_freq>\nFREQ_TABLE_KNEXT  \u003C\u002Fsysfs\u002F...\u002Favailable_frequencies>     # kernel-NEXT variant\nMAX_FREQ_KNEXT    \u003C\u002Fsysfs\u002F...\u002Fmax_freq>\nMIN_FREQ_KNEXT    \u003C\u002Fsysfs\u002F...\u002Fmin_freq>\n\n# 2. POWER_MODEL definitions — one block per profile\n\u003C POWER_MODEL ID=\u003Cint> NAME=\u003Cstring> >\nPARAM_NAME ARG_NAME \u003Cvalue>\n...\n\n# 3. PM_CONFIG — mandatory; selects boot default\n\u003C PM_CONFIG DEFAULT=\u003Cid> >\n```\n\nRules:\n\n- For `TYPE=FILE`, `\u003Cvalue>` is a **string** (`0`, `1`, `on`, `auto`, …).\n- For `TYPE=CLOCK`, `\u003Cvalue>` is an **integer** (Hz for clocks, raw integer for masks).\n- `-1` for a CLOCK value means **INT_MAX** (no cap).\n- The header `\u003C … >` line must start at column 0 with a space after `\u003C` and before `>`. Strict.\n- Every `PARAM_NAME` referenced in a POWER_MODEL must already be declared above.\n- **Frequency values must come from the kernel's `available_frequencies`** table for that clock — values not in the table get silently rounded.\n- `CORE_0` cannot be offlined; at least one CPU core must remain online in every profile.\n- **Copy PARAM names verbatim** from an existing POWER_MODEL block in the per-board file — chip families differ (T234 Orin uses `CPU_A78_\u003Cn>` for CPU clusters; T264 Thor uses a different convention). Don't invent.\n\n## Prerequisites\n\nResolve the active profile per\n[`..\u002F..\u002Fcontext\u002Ftarget-platform-contract.md`](..\u002F..\u002Fcontext\u002Ftarget-platform-contract.md).\nRefuse and route in these cases:\n\n| Condition | Refuse with |\n|---|---|\n| No active profile, or `active: NA` | Route to `\u002Fjetson-set-target` or `\u002Fjetson-init-target`. |\n| Profile lacks `bsp_image:` block | Route to `\u002Fjetson-init-image`. |\n| `\u003Cbsp_image.root_path>\u002FLinux_for_Tegra\u002F` missing | Route to `\u002Fjetson-init-image`. |\n| `\u003Csource.root_path>\u002FLinux_for_Tegra\u002F` missing or not a git repo | Route to `\u002Fjetson-init-source`. |\n\nResolve paths:\n\n- `\u003Cbsp_image.root_path>` from `bsp_image.root_path:` if present, else `\u003Cworkspace>\u002FImage`.\n- `\u003Csource.root_path>` from `source.root_path:` if present, else `\u003Cworkspace>\u002FSource`.\n\n`\u003Cbsp_image.root_path>` is **read-only** for this skill; every write lands\nunder `\u003Csource.root_path>` (the overlay tracker). This is the workflow\ninvariant in\n[`..\u002F..\u002Fcontext\u002Fbsp-customization-workflow.md#workflow-invariants`](..\u002F..\u002Fcontext\u002Fbsp-customization-workflow.md#workflow-invariants) —\nhand-editing upstream silently destroys the diff trail and makes\n`\u002Fjetson-promote-image` a noop.\n\n## The per-board file\n\nThe conf this skill edits has the relative path:\n\n```\nLinux_for_Tegra\u002Frootfs\u002Fetc\u002Fnvpmodel\u002Fnvpmodel_\u003Cactive-sku>.conf\n```\n\nIt lives in **two** roots; the skill walks both:\n\n| Role | Location | Skill writes? |\n|---|---|---|\n| Detection + pristine source | `\u003Cbsp_image.root_path>\u002FLinux_for_Tegra\u002Frootfs\u002Fetc\u002Fnvpmodel\u002F` | no — read-only |\n| Overlay edit target + git commit | `\u003Csource.root_path>\u002FLinux_for_Tegra\u002Frootfs\u002Fetc\u002Fnvpmodel\u002F` | yes |\n\nSubsequent sections refer to **the per-board file** to mean the overlay copy\nunder `\u003Csource.root_path>`. Operations 1–4 all read, edit, and save against\nthat overlay copy. The `\u003Cbsp_image.root_path>` copy is read once during the\n\"Resolving `\u003Cactive-sku>`\" detection step and once during the\n[Overlay edit recipe](#overlay-edit-recipe-apply-before-any-operation)'s\npristine-import step, then never touched again.\n\n### Resolving `\u003Cactive-sku>` — which file to edit\n\nThe filename is **not** always `module.id + \"_\" + module.sku`. Variants exist:\n\n- `nvpmodel_p3767_0000_super.conf` (super-mode SKU variant)\n- `nvpmodel_igx_orin.conf`, `nvpmodel_igx_orin_safety.conf` (IGX, no SKU number)\n\nAt boot, `nvpower.sh` (at `Linux_for_Tegra\u002Frootfs\u002Fetc\u002Fsystemd\u002Fnvpower.sh`) reads the kernel DTB's root `compatible` string and maps it — plus super\u002Fsafety state — to the nvpmodel filename. Replicate that mapping **BSP-side**, against `\u003Cbsp_image.root_path>`:\n\n1. Resolve the SKU-correct kernel DTB under `\u003Cbsp_image.root_path>\u002FLinux_for_Tegra\u002F` and read its root `compatible` (detect kernel DTB from the active flash conf).\n2. Read `\u003Cbsp_image.root_path>\u002FLinux_for_Tegra\u002Frootfs\u002Fetc\u002Fsystemd\u002Fnvpower.sh` to find the compatible-string → conf mapping, and apply it to the compatible from the previous step (factoring in super\u002Fsafety flags).\n3. Verify the resolved `nvpmodel_\u003C...>.conf` exists under `\u003Cbsp_image.root_path>\u002FLinux_for_Tegra\u002Frootfs\u002Fetc\u002Fnvpmodel\u002F`.\n\nShortcut: filter `\u003Cbsp_image.root_path>\u002FLinux_for_Tegra\u002Frootfs\u002Fetc\u002Fnvpmodel\u002F` to `nvpmodel_\u003Cmodule.id>_\u003Cmodule.sku>*.conf`; for super-mode flash configs pick the `_super` variant. Use only when unambiguous; otherwise fall back to the DTB method.\n\nDon't blindly compose `nvpmodel_\u003Cid>_\u003Csku>.conf` — verify the file actually exists.\n\n### Propagation set — confs to keep in sync\n\nThe active-SKU file is rarely the only conf that should carry a customization. After editing it, **apply the same edit to every sibling in the propagation set** so the change survives regardless of which module \u002F baseboard SKU is booted:\n\n- **The reference platform's nvpmodel conf** — the upstream conf the active file was forked from (resolve via `reference_devkit` in the active profile, or `jetson-derive-carrier` fork ancestry). For a BSP that contains only the reference (no derived carriers), this is the same file as the active and the rule reduces to a no-op.\n- **Every carrier-derived nvpmodel conf** — each `nvpmodel_*.conf` produced by `jetson-derive-carrier` for a custom carrier on top of the same module SKU.\n- **`_super` siblings** when present (e.g. `nvpmodel_p3767_0000_super.conf`) — apply the structural edit (new POWER_MODEL block, removed block, `PM_CONFIG DEFAULT=` flip, NAME rename) **but preserve the super conf's higher MAX_FREQ \u002F MIN_FREQ caps**. Those elevated caps are the whole reason the `_super` variant exists; a blanket content overwrite from the non-super file would silently flatten the super envelope.\n\n**\"Apply the same edit\" ≠ blanket file copy.** Port the changed POWER_MODEL \u002F PARAM lines into each sibling; preserve every other line. Blanket-copying is safe **only** when both confs were byte-identical before the edit. Otherwise re-validate per the Rules on each target (available-frequencies table for clock values, ID uniqueness, `PM_CONFIG DEFAULT=` references a present ID) — sibling confs may carry SKU-specific `available_frequencies` tables that don't accept the active file's frequency values.\n\n## Overlay edit recipe (apply before any Operation)\n\nFollow the canonical\n[Off-skill edits recipe](..\u002F..\u002Fcontext\u002Fbsp-customization-workflow.md#off-skill-edits)\nin the workflow doc — pristine import + customization commit pair, both\ngated by the preview gate. Apply once per run, covering every per-board\nfile the run touches (the active conf plus every sibling in the\n[Propagation set](#propagation-set--confs-to-keep-in-sync)).\n\nConcrete substitutions for this skill:\n\n- `\u003Crel>\u002F\u003Cfile>` is `rootfs\u002Fetc\u002Fnvpmodel\u002F\u003Cconf>`.\n- Suggested pristine-import message:\n  `import pristine: \u003Ccomma-separated rel paths of imported confs>`,\n  body `Source: \u003Cbsp_image.root_path>\u002FLinux_for_Tegra\u002F (BSP \u003Cbsp_image.version>)`.\n- Suggested customization-commit header:\n  `jetson-customize-nvpmodel: \u003Csummary>`,\n  body lines like `nvpmodel_p3767_0001.conf: PM_CONFIG DEFAULT 2 -> 0 (MAXN)`.\n\n## Instructions\n\nPick the operation that matches the user's intent and follow the\nmatching subsection. All write-side operations (1–4) must first\napply the [Overlay edit recipe](#overlay-edit-recipe-apply-before-any-operation).\n\n- Operation 1 — Add a new power mode (drives the NVIDIA Power Estimator).\n- Operation 2 — Remove a power mode.\n- Operation 3 — Edit an existing power mode (clamps, core-online, NAME).\n- Operation 4 — Change the boot default (`PM_CONFIG DEFAULT=`).\n- Operation 5 — List defined power modes (read-only).\n\nAfter any write-side operation, run the Deploy chain (`## Deploy`)\nto land the change on the device.\n\n## Examples\n\nAdd a new 30 W power mode from a Power Estimator export and pin it\nas the boot default on a P3767-0001 target:\n\n```\n\u002Fjetson-customize-nvpmodel\n> add a new POWER_MODEL from ~\u002FDownloads\u002Fnvpmodel_30W.conf as ID 8 NAME \"30W_CUSTOM\"\n> set PM_CONFIG DEFAULT to 8\n```\n\nCap a Thor target to the MAXN power envelope by flipping the boot\ndefault (no envelope tuning needed):\n\n```\n\u002Fjetson-customize-nvpmodel\n> set PM_CONFIG DEFAULT to the MAXN profile's ID\n```\n\nList which power modes the active SKU currently defines and which\nis the boot default:\n\n```\n\u002Fjetson-customize-nvpmodel\n> list defined power modes\n```\n\n## Operation 1 — Add a new power mode\n\nApply the [Overlay edit recipe](#overlay-edit-recipe-apply-before-any-operation) first.\n\n### Generate the profile via the Power Estimator\n\n**Must not interpolate or extrapolate frequencies from existing profiles to estimate power.**\n\nFor any non-stock power envelope (custom budget, custom workload), use the **NVIDIA Power Estimator**: `https:\u002F\u002Fjetson-tools.nvidia.com\u002Fpowerestimator\u002F`\n\n- Pick the exact module SKU and JetPack release.\n- Enter workload (CPU cores, GPU usage, EMC, codecs, camera, display).\n- Estimate power budget.\n- Download custom `nvpmodel.conf`.\n- Share the downloaded `nvpmodel.conf` file path.\n\n### Append the POWER_MODEL block\n\nIn the per-board file, insert the new block **after** the last `\u003C PARAM … >` declaration (POWER_MODEL must reference declared PARAMs) and **before** the final `\u003C PM_CONFIG ... >` line. The block structure is the **File format** shown above; frequency values come from the Power Estimator output (see \"Generate the profile via the Power Estimator\").\n\nCheck `ID` is unique; `NAME` should be uppercase, no whitespace. Each numeric `MAX_FREQ` \u002F `MIN_FREQ` value must satisfy the `available_frequencies` rule (see Rules). If you intend this mode to be the boot default, follow Operation 4.\n\n## Operation 2 — Remove a power mode\n\nApply the [Overlay edit recipe](#overlay-edit-recipe-apply-before-any-operation) first.\n\n1. In the per-board file, delete the entire `\u003C POWER_MODEL ID=\u003Cn> NAME=... >` block including all its parameter lines, up to (but not including) the next `\u003C POWER_MODEL ... >` or `\u003C PM_CONFIG ... >` marker.\n2. If the deleted ID matches the current `DEFAULT=\u003Cid>` value in the trailing `\u003C PM_CONFIG … >` line, point `DEFAULT=` at a remaining ID — otherwise nvpmodel will fail to apply a default at boot.\n3. Search for hard-coded references in the rootfs scripts before declaring the change safe:\n   ```bash\n   grep -rn \"nvpmodel -m\" \\\n     Linux_for_Tegra\u002Frootfs\u002Fetc \\\n     Linux_for_Tegra\u002Frootfs\u002Fopt 2>\u002Fdev\u002Fnull\n   ```\n\nID gaps are legal — you don't have to renumber remaining modes.\n\n## Operation 3 — Edit an existing power mode\n\nApply the [Overlay edit recipe](#overlay-edit-recipe-apply-before-any-operation) first.\n\n1. In the per-board file, edit the parameter lines inside the block. Keep `\u003Cparam>` and `\u003Carg>` names exactly matching the PARAM declarations.\n2. **If the edit changes the power envelope** (any CPU\u002FGPU\u002FEMC\u002FPVA\u002FDLA `MAX_FREQ` \u002F `MIN_FREQ`, core-online count for a freq-bound mode, or TPC mask), re-run the **NVIDIA Power Estimator** (see \"Generate the profile via the Power Estimator\") to ground the new frequencies in a real per-component model. Do not interpolate from neighboring modes.\n3. Validate clock values per the `available_frequencies` rule (see Rules).\n\nEdits that only toggle core-online flags within an already-validated envelope, or only change `NAME=`, don't need the Power Estimator pass.\n\n## Operation 4 — Change the boot default\n\nApply the [Overlay edit recipe](#overlay-edit-recipe-apply-before-any-operation) first.\n\nEdit the `\u003C PM_CONFIG DEFAULT=\u003Cid> >` line at the bottom of the per-board file.\n\n`\u003Cid>` must reference an existing `\u003C POWER_MODEL ID=\u003Cid> … >` block in the same file. Pointing at an undefined ID makes nvpmodel fail to apply at boot.\n\n## Operation 5 — List defined power modes\n\nThis is a read-only operation; no overlay-tracker setup is needed. Run\nagainst whichever copy you want to inspect (`\u003Cbsp_image.root_path>\u002F...`\nfor the pristine state, `\u003Csource.root_path>\u002F...` for the post-edit state):\n\n```bash\ngrep -E '^\u003C POWER_MODEL ' \u003Cper-board file>\ngrep -E '^\u003C PM_CONFIG '   \u003Cper-board file>\n```\n\nThe first command prints every `ID`\u002F`NAME`; the second prints the current boot default. On a running target, `nvpmodel -p --verbose` (or `nvpmodel -q`) is authoritative.\n\n## Limitations\n\n- BSP-side scope only — this skill never invokes `nvpmodel -m` on a\n  running target. Live mode switching requires reboot via Deploy,\n  or the side-channel `scp + nvpmodel -m` flow described in `## Deploy`.\n- Edits land in the overlay copy under `\u003Csource.root_path>` only;\n  the `\u003Cbsp_image.root_path>` copy is read-only and is rewritten by\n  `\u002Fjetson-promote-image`. Hand-editing `bsp_image` is silently lost\n  on the next `\u002Fjetson-init-image` re-extract.\n- Frequency values must come from the kernel's `available_frequencies`\n  table for the relevant clock — values outside the table get\n  silently rounded. This skill does not validate against a live\n  target's table; trust the per-board file's existing values and\n  Power Estimator output.\n- Non-trivial envelope edits (any CPU\u002FGPU\u002FEMC\u002FPVA\u002FDLA `MAX_FREQ` \u002F\n  `MIN_FREQ` change) require the NVIDIA Power Estimator\n  (`https:\u002F\u002Fjetson-tools.nvidia.com\u002Fpowerestimator\u002F`) — interpolation\n  or extrapolation between existing modes is not supported.\n- Propagation across siblings is partial by design: only the changed\n  `POWER_MODEL` \u002F `PARAM` lines are ported, never a blanket file\n  overwrite, since `_super` siblings carry elevated `MAX_FREQ`\u002F`MIN_FREQ`\n  caps that must stay intact.\n- PARAM names are chip-family-specific (e.g. `CPU_A78_\u003Cn>` on T234,\n  different on T264). Copy verbatim from existing POWER_MODEL blocks;\n  invented names silently fail to apply.\n\n## Troubleshooting\n\n| Error | Cause | Solution |\n|---|---|---|\n| nvpmodel fails to apply default at boot | `PM_CONFIG DEFAULT=\u003Cid>` references a deleted or missing POWER_MODEL ID | Point `DEFAULT=` at an existing `\u003C POWER_MODEL ID=\u003Cn> ... >` in the same file. |\n| Frequency value silently doesn't take effect | Value not in the kernel's `available_frequencies` table for that clock | Replace with the nearest legal value from the running target's `available_frequencies` (or Power Estimator output). |\n| `MAX_FREQ -1` does nothing | `-1` on a `TYPE=FILE` PARAM (only valid for `TYPE=CLOCK`) | Use `-1` only on CLOCK params; for FILE params, write the actual string the sysfs node expects. |\n| Module dies \u002F fails to boot after edit | `CORE_0` was offlined, or the mode dropped below the platform's minimum quiescent envelope | Keep `CORE_0` online; floor `MIN_FREQ` per the Power Estimator's lowest-power profile for the SKU. |\n| `nvpmodel -m \u003Cid>` returns \"ID not found\" at runtime | Mode ID was removed or renumbered but a rootfs script still references the old ID | `grep -rn 'nvpmodel -m' rootfs\u002Fetc rootfs\u002Fopt` and update \u002F remove the stale references. |\n| Change vanished after `\u002Fjetson-init-image` re-extract | Edit landed in `\u003Cbsp_image.root_path>` instead of `\u003Csource.root_path>` overlay | Re-apply via the [Overlay edit recipe](#overlay-edit-recipe-apply-before-any-operation) so the change is committed in the overlay tracker. |\n| `_super` sibling lost elevated caps after propagation | Blanket file copy clobbered `_super`'s higher `MAX_FREQ`\u002F`MIN_FREQ` | Port only the structural change; preserve `_super`'s caps per [Propagation set](#propagation-set--confs-to-keep-in-sync). |\n| Parser rejects new POWER_MODEL block | Header `\u003C … >` not at column 0, missing space after `\u003C` or before `>`, or PARAM not declared earlier | Restore strict header formatting; ensure every referenced `PARAM_NAME` appears in a `\u003C PARAM ... >` block above. |\n\n## Deploy\n\nThe customization commit in the overlay tracker does not reach the device\non its own. The Deploy chain:\n\n1. **`\u002Fjetson-promote-image`** — copies every tracked file in the overlay\n   into `\u003Cbsp_image.root_path>\u002FLinux_for_Tegra\u002F`. Diff-aware (skip\n   byte-identical); uses `sudo cp -p` for `rootfs\u002F*` destinations.\n2. **`\u002Fjetson-flash-image`** — flashes the updated `bsp_image` to the\n   device.\n3. (Alternate, no flash) Copy `\u003Csource.root_path>\u002FLinux_for_Tegra\u002Frootfs\u002Fetc\u002Fnvpmodel\u002F\u003Cconf>`\n   directly to the running target's `\u002Fetc\u002Fnvpmodel\u002F\u003Cconf>`, then\n   `sudo nvpmodel -m \u003Cid>` (or reboot to pick up the new `DEFAULT=`).\n\nEditing `\u003Csource.root_path>\u002F...` without committing — or editing\n`\u003Cbsp_image.root_path>\u002F...` directly — does nothing for `\u002Fjetson-promote-image`\nand is silently lost on the next `\u002Fjetson-init-image` re-extract.\n",{"name":5,"license":14,"version":23,"metadata":24,"description":7},"0.0.1",{"tags":25,"author":29,"domain":26,"data-classification":30},[26,27,28],"power","nvpmodel","profile","Jetson Team","public","main","fc4f2c7759ae400c66cd38656db70ead882a0d08","catalog\u002Fbundles\u002Fnvidia\u002Fjetson-customize-nvpmodel","https:\u002F\u002Fgithub.com\u002Fmatheussluzz\u002FZeuZ-Agent","https:\u002F\u002Fgithub.com\u002Fmatheussluzz\u002FZeuZ-Agent#readme","https:\u002F\u002Fgithub.com\u002Fmatheussluzz","2026-08-27 05:21:51",{"source":8,"skillName":5,"commands":39,"prompt":49,"zipAvailable":50},[40,43,46],{"tool":41,"command":42},"npx","npx skills add matheussluzz\u002FZeuZ-Agent --skill jetson-customize-nvpmodel",{"tool":44,"command":45},"bunx","bunx skills add matheussluzz\u002FZeuZ-Agent --skill jetson-customize-nvpmodel",{"tool":47,"command":48},"pnpm","pnpm dlx skills add matheussluzz\u002FZeuZ-Agent --skill jetson-customize-nvpmodel","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fmatheussluzz\u002FZeuZ-Agent\u002Fblob\u002Fmain\u002Fcatalog\u002Fbundles\u002Fnvidia\u002Fjetson-customize-nvpmodel\u002FSKILL.md",true,[52,58,62,68],{"path":53,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":56,"sourceHtmlUrl":57},"BENCHMARK.md","REFERENCE","参考文档",3587,"https:\u002F\u002Fgithub.com\u002Fmatheussluzz\u002FZeuZ-Agent\u002Fblob\u002Fmain\u002Fcatalog\u002Fbundles\u002Fnvidia\u002Fjetson-customize-nvpmodel\u002FBENCHMARK.md",{"path":59,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":60,"sourceHtmlUrl":61},"skill-card.md",3673,"https:\u002F\u002Fgithub.com\u002Fmatheussluzz\u002FZeuZ-Agent\u002Fblob\u002Fmain\u002Fcatalog\u002Fbundles\u002Fnvidia\u002Fjetson-customize-nvpmodel\u002Fskill-card.md",{"path":63,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":66,"sourceHtmlUrl":67},"skill.oms.sig","OTHER","其他",4601,"https:\u002F\u002Fgithub.com\u002Fmatheussluzz\u002FZeuZ-Agent\u002Fblob\u002Fmain\u002Fcatalog\u002Fbundles\u002Fnvidia\u002Fjetson-customize-nvpmodel\u002Fskill.oms.sig",{"path":69,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":70,"sourceHtmlUrl":71},"zeuz.manifest.yaml",172,"https:\u002F\u002Fgithub.com\u002Fmatheussluzz\u002FZeuZ-Agent\u002Fblob\u002Fmain\u002Fcatalog\u002Fbundles\u002Fnvidia\u002Fjetson-customize-nvpmodel\u002Fzeuz.manifest.yaml",[73,81,88,95,102,109],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":77,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":18},"01M78JQ2R0QR8H2Z6662RFX804","gamma","Fixture skill gamma for discovery benchmarks.","MIT","https:\u002F\u002Fgithub.com\u002Fmatheussluzz\u002FZeuZ-Agent\u002Fblob\u002Fmain\u002Ftest\u002Ffixtures\u002Fskill-registry-mini\u002Fskills\u002Fgamma\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":77,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":18},"01M78JQ1KCVZ15APZG3FFF29K0","beta","Fixture skill beta for discovery benchmarks.","https:\u002F\u002Fgithub.com\u002Fmatheussluzz\u002FZeuZ-Agent\u002Fblob\u002Fmain\u002Ftest\u002Ffixtures\u002Fskill-registry-mini\u002Fskills\u002Fbeta\u002FSKILL.md",[],[],{"skillNo":89,"name":90,"nameZh":6,"description":91,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":77,"licenseFileUrl":6,"sourceHtmlUrl":92,"categories":93,"keywords":94,"publishedAt":18},"01M78JQ0FJTK4Q9JKZMX9SVSEN","alpha","Fixture skill alpha for discovery benchmarks.","https:\u002F\u002Fgithub.com\u002Fmatheussluzz\u002FZeuZ-Agent\u002Fblob\u002Fmain\u002Ftest\u002Ffixtures\u002Fskill-registry-mini\u002Fskills\u002Falpha\u002FSKILL.md",[],[],{"skillNo":96,"name":97,"nameZh":6,"description":98,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":77,"licenseFileUrl":6,"sourceHtmlUrl":99,"categories":100,"keywords":101,"publishedAt":18},"01M78JPYVYPY7R8QS10KADQPWE","writing-for-agents","Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.","https:\u002F\u002Fgithub.com\u002Fmatheussluzz\u002FZeuZ-Agent\u002Fblob\u002Fmain\u002Fskills\u002Fwriting-for-agents\u002FSKILL.md",[],[],{"skillNo":103,"name":104,"nameZh":6,"description":105,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":77,"licenseFileUrl":6,"sourceHtmlUrl":106,"categories":107,"keywords":108,"publishedAt":18},"01M78JPVYHB88RF5ATZVGX84RR","wizard","Generate an interactive bash wizard that walks a human through steps only they can perform. Use when provisioning infrastructure, setting up credentials or CI secrets, walking an unfamiliar third-party dashboard, or running a one-off migration or cutover. Don't invoke this for steps the agent can perform itself.","https:\u002F\u002Fgithub.com\u002Fmatheussluzz\u002FZeuZ-Agent\u002Fblob\u002Fmain\u002Fskills\u002Fwizard\u002FSKILL.md",[],[],{"skillNo":110,"name":111,"nameZh":6,"description":112,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":77,"licenseFileUrl":6,"sourceHtmlUrl":113,"categories":114,"keywords":115,"publishedAt":18},"01M78JPSSWNTZW89DWGJTX7716","wayfinder","Plan a huge chunk of work (more than one agent session can hold) as a shared map of decision tickets on your issue tracker, and resolve them one at a time until the way to the destination is clear.","https:\u002F\u002Fgithub.com\u002Fmatheussluzz\u002FZeuZ-Agent\u002Fblob\u002Fmain\u002Fskills\u002Fwayfinder\u002FSKILL.md",[],[]]