comfy-node
Orchestrate a ComfyUI node pack from idea to registry: scaffold, create + seed the repo, open the gitops adoption PR. Use when releasing or spinning up a new comfyui node pack.
What this skill does
# comfy-node
Take a ComfyUI custom-node **idea** and drive it through every step from empty
to publish-ready, collapsing the manual repo-creation + gitops wiring into one
orchestrated pass with a single human approval gate.
This is the orchestrator around the `comfyui-node-scaffold` skill (which only
generates the local repo). Use `comfyui-node-scaffold` alone if you just want
the files; use **this** when you want the GitHub repo created, pushed, and
adopted into gitops too.
## When to Use This Skill
| Use this skill when... | Use the alternative when... |
|---|---|
| The user gives an idea and wants the whole pipeline stood up — repo created, seeded, and gitops-adopted | You only want the local files → `comfyui-node-scaffold` |
| Spinning up a `project:comfyui-nodes` taskwarrior backlog idea end to end | Adding a node to an *existing* pack → edit that repo directly |
Do **not** use it to add a node to an existing pack, or to publish a release
(release-please + `publish.yml` already automate that once the repo exists).
## The shape it automates
```mermaid
flowchart LR
idea["idea"] --> sc["scaffold.py"] --> gh["gh repo create<br/>+ seed main"] --> gop["gitops PR<br/>(entry + import block)"]
gop --> gate["👤 merge gitops PR"] --> apply["Scalr apply:<br/>adopt + secrets + protection"] --> rm["remove import block"] --> impl["implement + release"]
classDef g fill:#1b4332,stroke:#2d6a4f,color:#fff
classDef m fill:#6a040f,stroke:#9d0208,color:#fff
class sc,gh,gop,apply,rm g
class gate,impl m
```
Everything left of the gate is one orchestrated pass. There is **no scaffold
PR** — the seed goes straight to `main` (see Phase 3 for why). The single gate
(merging the gitops PR) is intentionally human — it triggers an infra `apply`
on shared state. Never merge it on the user's behalf.
## Preconditions
- Run from the workspace root `repos/laurigates/` (new repo lands as a sibling
of the reference packs; modal-variant primitive copy resolves there).
- `gh auth status` is a **personal** account that can create repos. The gitops
GitHub App *cannot* create repos on user accounts — that is exactly why the
repo is created out-of-band here and then imported into Terraform state.
- The gitops repo is clean (no uncommitted `repositories.tf` / `main.tf`
changes) so the orchestrator's gitops PR is isolated.
## Phase 0 — Derive and confirm the spec
From the idea, derive and **show the user** before creating anything external:
| Field | How to derive | Example |
|-------|---------------|---------|
| `--name` | `comfyui-<kebab>`; reuse the family prefix (`touch-…` for touch UX). | `comfyui-touch-resize` |
| `--display` | Title-case. | `Touch Resize` |
| `--desc` | One line, registry-facing. | `Selection-gated pinch-to-resize for ComfyUI nodes and groups on touch devices.` |
| `--variant` | `gesture` for canvas interactions (resize/move/region); `frontend` for a per-widget modal; `backend` only if it reads disk / serves data. | `gesture` |
| `--widgets` | CSV of target widget names (modal variants only; omit for `gesture`). | — |
| topics | `["comfyui","comfyui-nodes",…]` + facet tags. | `…,"mobile","touch","resize"` |
Confirm the name and variant with the user — these are hard to change after the
repo exists. If the idea matches a `project:comfyui-nodes` task, mark it
in_progress:
```sh
task project:comfyui-nodes export | jq -r '.[] | select(.description | test("resize"; "i")) | .uuid'
```
## Phase 1 — Preflight (fail fast if the name is taken)
```sh
test ! -e comfyui-touch-resize && echo "local: free" || echo "local: EXISTS"
```
```sh
grep -q '"comfyui-touch-resize"' gitops/repositories.tf && echo "gitops: EXISTS" || echo "gitops: free"
```
```sh
gh repo view laurigates/comfyui-touch-resize >/dev/null 2>&1 && echo "github: EXISTS" || echo "github: free"
```
All three must report free. Stop and surface any collision.
## Phase 2 — Scaffold + local green check
```sh
python3 .claude/skills/comfyui-node-scaffold/scaffold.py --name comfyui-touch-resize --display "Touch Resize" --desc "Selection-gated pinch-to-resize for ComfyUI nodes and groups on touch devices." --variant gesture
```
Then bring the pack to green locally (the scaffold prints these too):
```sh
cd comfyui-touch-resize
```
```sh
uv sync --group dev
```
```sh
npm install --no-audit --no-fund
```
```sh
just check
```
`just check` must pass before anything is pushed. If it fails, fix locally and
re-run — do not create the remote repo on a red pack.
## Phase 3 — Create the GitHub repo and seed `main`
Seed `main` **directly** as the first commit — no scaffold branch, no PR. The
repo has no branch protection yet (gitops adds it on adoption in Phase 5), so
this is allowed, and it avoids the branch juggling you'd otherwise hit: if the
first push were a feature branch, `main` would be missing on origin, forcing a
later rename + default-branch change + base-branch fixups. Pushing `main` first
sidesteps all of it. Implementation work afterward goes through feature-branch
PRs as normal (protection is live by then).
```sh
git init -b main
```
```sh
git add -A
```
```sh
git commit -m "feat: scaffold comfyui-touch-resize (gesture pack)"
```
```sh
gh repo create laurigates/comfyui-touch-resize --public --source . --remote origin --push
```
> **Branch-protection hook note (expect this):** in a Claude Code session the
> `branch-protection` hook **will** block the agent from `git add`/`commit` on
> `main` (confirmed on the first real run). For a brand-new, not-yet-protected
> repo this is a false positive. Hand the whole seed to the user as one
> paste-safe line to run with the `! ` prefix, e.g.:
>
> ```
> cd <repo> && git add -A && git commit -m "feat: scaffold <name> (gesture pack)" && gh repo create laurigates/<name> --public --source . --remote origin --push
> ```
>
> (`git add -A` and `&&`-chaining are fine in the *user's* shell — those hooks
> are agent-side.) Do **not** work around it by seeding a feature branch — that
> reintroduces the missing-`main`/rename juggling this phase exists to avoid.
> Don't fight the hook with quoting tricks either.
The `--push` makes the seeded `main` the default branch.
## Phase 4 — Open the gitops PR (entry + transient import block)
Two edits in the `gitops/` repo, on a dedicated branch.
**`gitops/repositories.tf`** — add to the active repositories `locals` block,
next to the other `comfyui-*` entries (mirror `comfyui-touch-connect`):
```hcl
"comfyui-touch-resize" = {
description = "Selection-gated pinch-to-resize for ComfyUI nodes and groups on touch devices"
visibility = "public"
release_please = true
comfy_registry = true
topics = ["comfyui", "comfyui-nodes", "mobile", "touch", "resize"]
}
```
**`gitops/main.tf`** — add a transient `import` block alongside the existing
ones at the top of the file:
```hcl
import {
to = github_repository.this["comfyui-touch-resize"]
id = "comfyui-touch-resize"
}
```
Validate, branch, commit, push, open the PR (run inside `gitops/`):
```sh
just check
```
```sh
git -C gitops switch -c feat/adopt-comfyui-touch-resize
```
```sh
git -C gitops add repositories.tf main.tf
```
```sh
git -C gitops commit -m "feat: adopt comfyui-touch-resize (comfy_registry)"
```
```sh
git -C gitops push -u origin feat/adopt-comfyui-touch-resize
```
```sh
gh pr create -R laurigates/gitops -a laurigates -l chore -l opentofu --title "feat: adopt comfyui-touch-resize (comfy_registry)" --body-file /tmp/gitops-pr-body.md
```
Write a short body (to `/tmp/gitops-pr-body.md`) rather than `--fill` — it's an
infra PR that triggers an apply, so spell out what merge does: imports the repo,
pushes `REGISTRY_ACCESS_TOKEN` + release-please credentials, applies the
branch-protection ruleset, and that a follow-up PR removes the import block. Use
labels `chore` + `opentofu` (both exist in the gitops repo; check
`gh label list -R laurigates/gitops` if unsure).
Set metadata per `gitRelated in Backend & APIs
jfrog
IncludedInteract with the JFrog Platform via the JFrog CLI and REST/GraphQL APIs. Use this skill when the user wants to manage Artifactory repositories, upload or download artifacts, manage builds, configure permissions, manage users and groups, work with access tokens, configure JFrog CLI servers, search artifacts, manage properties, set up replication, manage JFrog Projects, run security audits or scans, look up CVE details, query exposures scan results from JFrog Advanced Security, manage release bundles and lifecycle operations, aggregate or export platform data, or perform any JFrog Platform administration task. Also use when the user mentions jf, jfrog, artifactory, xray, distribution, evidence, apptrust, onemodel, graphql, workers, mission control, curation, advanced security, exposures, or any JFrog product name.
cupynumeric-migration-readiness
IncludedPre-migration readiness assessor for porting NumPy to cuPyNumeric. Use BEFORE substantial porting work begins when the user asks whether code will scale on GPU, whether they should migrate to cuPyNumeric, which NumPy patterns transfer cleanly, what must be refactored before porting, or mentions pre-port assessment, scaling analysis, or refactor planning. Inspect the user's source code, look up NumPy usage, cross-reference the cuPyNumeric API support manifest, and distinguish distributed-scaling-friendly patterns from blockers such as unsupported APIs, scalar synchronization, host round-trips, Python/object-heavy control flow, shape/data-dependent branching, and in-place mutation hazards. Produce a verdict of READY, LIGHT REFACTOR, SIGNIFICANT REFACTOR, or NOT RECOMMENDED, with concrete refactor pointers.
alibabacloud-data-agent-skill
IncludedInvoke Alibaba Cloud Apsara Data Agent for Analytics via CLI to perform natural language-driven data analysis on enterprise databases. Data Agent for Analytics is an intelligent data analysis agent developed by Alibaba Cloud Database team for enterprise users. It automatically completes requirement analysis, data understanding, analysis insights, and report generation based on natural language descriptions. This tool supports: discovering data resources (instances/databases/tables) managed in DMS, initiating query or deep analysis sessions, real-time progress tracking, and retrieving analysis conclusions and generated reports. Use this Skill when users need to query databases, analyze data trends, generate data reports, ask questions in natural language, or mention "Data Agent", "data analysis", "database query", "SQL analysis", "data insights".
token-optimizer
IncludedReduce OpenClaw token usage and API costs through smart model routing, heartbeat optimization, budget tracking, and native 2026.2.15 features (session pruning, bootstrap size limits, cache TTL alignment). Use when token costs are high, API rate limits are being hit, or hosting multiple agents at scale. The 4 executable scripts (context_optimizer, model_router, heartbeat_optimizer, token_tracker) are local-only — no network requests, no subprocess calls, no system modifications. Reference files (PROVIDERS.md, config-patches.json) document optional multi-provider strategies that require external API keys and network access if you choose to use them. See SECURITY.md for full breakdown.
resend-cli
IncludedUse this skill when the task is specifically about operating Resend from an AI agent, terminal session, or CI job via the official resend CLI: installing/authenticating the CLI, sending/listing/updating/cancelling emails, batch sends, domains and DNS, webhooks and local listeners, inbound receiving, contacts, topics, segments, broadcasts, templates, API keys, profiles, or debugging Resend CLI/API failures. Trigger on mentions of Resend CLI, `resend`, `resend doctor`, `resend emails send`, `resend domains`, `resend webhooks listen`, `resend emails receiving`, or agent-friendly terminal automation.
alibabacloud-odps-maxframe-coding
IncludedUse this skill for MaxFrame SDK development and documentation navigation on Alibaba Cloud MaxCompute (ODPS). Helps answer MaxFrame API, concept, official example, and supported pandas API questions; create data processing programs; read/write MaxCompute tables; debug jobs (remote or local); and build custom DPE runtime images. Trigger when users mention MaxFrame, MaxCompute with MaxFrame, ODPS table processing, DPE runtime, MaxFrame docs/examples, DataFrame/Tensor operations, or GPU runtime setup. Works for both English and Chinese queries about Alibaba Cloud data processing with MaxFrame.