Develop

AGENTS.md

@@ -5,18 +5,23 @@ These rules apply to AI-assisted work in this repository. `CONTRIBUTING.md` is t
 ## Scope And Runtime Boundary
 
 - Keep each change focused on a diagnosed problem. Trace the existing call path and tests before editing, and remove incidental generated files from the diff.
-- Hugging Face Diffusers and Modular Diffusers are MoDiff's only supported model-execution layer. Do not add an alternate graph executor, hosted inference provider, independent Transformers application, or another model driver.
-- Supporting libraries used by Diffusers and ordinary deterministic media processing are not alternate drivers. They must remain narrowly scoped, documented, and covered by tests.
+- MoDiff may execute models through an official library maintained and published by Hugging Face. A Hub repository, organization name, or compatible API is not enough: verify the library's upstream ownership and package provenance, then declare and review the exact integration in MoDiff's executable dependency contract.
+- All supported libraries run locally behind MoDiff's existing node graph, resource management, Auto/Expert, file, and security boundaries. Do not add an alternate graph executor, hosted inference provider, browser-side runtime, or library-owned workflow representation.
+- Keep nodes and client contracts task- or modality-generic. Library- and model-specific loading, parameter aliases, and output normalization belong in small backend adapters selected from a declared execution specification, not new model-named nodes or frontend branches.
+- Transformers is an optional runtime, not a default application dependency. Do not install it during startup, registry discovery, template browsing/opening, or Auto planning. A workflow that requires it must declare a reviewed optional runtime profile, present an explicit install/consent action, and verify the installed version before becoming runnable.
+- Ordinary deterministic media processing is allowed when narrowly scoped, documented, and covered by tests.
 - Never enable arbitrary remote Python code, mutable model revisions, or custom model execution implicitly. Trust-sensitive behavior requires an explicit operator choice and an immutable revision.
 - Prefer existing module, graph, configuration, error, and test patterns. Do not create a parallel workflow representation or model-loading path.
 
-## Diffusers And Modular Diffusers
+## Hugging Face Model Libraries
 
 - Keep the reviewed Diffusers revision pinned in the executable installation contract and update its compatibility test when changing it.
+- Treat support for each additional official Hugging Face library as an explicit integration: document its purpose and provenance, constrain its compatible version, keep heavyweight imports lazy, and add no-download compatibility and boundary tests.
+- Official-library eligibility is not blanket trust for Hub artifacts or repository code. Prefer `safetensors`; pin curated models and adapters immutably; and require a separate explicit operator decision for any reviewed remote-code path.
 - Use Diffusers loaders, pipelines, components, schedulers, adapters, and offload hooks instead of reimplementing upstream behavior.
 - Modular blocks should declare inputs, outputs, and dependencies clearly, avoid hidden cross-block state, and remain composable through `init_pipeline`.
 - Keep model-specific differences explicit and small. Put reusable behavior in the existing shared Diffusers modules rather than copying it into another pipeline.
-- Prefer `safetensors`; document and test any unavoidable unsafe deserialization or remote-code boundary.
+- For non-Diffusers libraries, put reusable behavior in the corresponding generic task module and preserve the same graph, resource, progress, cancellation, and output contracts.
 
 ## Security And Data
 

CONTRIBUTING.md

@@ -65,6 +65,10 @@ uv pip check --python .venv/bin/python
 
 Use the corresponding accelerator instead of `cpu` when the change affects CUDA, ROCm, or MPS. Update the compatibility manifest and public installation guidance only when the evidence supports the claim. MoDiff deliberately has no `uv.lock`; do not generate one or describe the top-level requirements files as a cross-platform lock.
 
+MoDiff may integrate model libraries officially maintained and published by Hugging Face, but each library remains a separately reviewed execution dependency. Verify its upstream ownership, package provenance, license, supported version, loading behavior, and remote-code boundary. Hub hosting alone does not establish that a library or model is maintained by Hugging Face. Keep execution in the existing backend graph and expose task-generic contracts to the client.
+
+Transformers-specific execution is opt-in. Do not add Transformers to the base application environment or install an optional runtime merely because a template is viewed, nodes are discovered, or Auto compatibility is planned. A requiring workflow must identify its versioned optional runtime, show an explicit install/consent action, perform installation outside graph execution, and re-run package and compatibility checks before the workflow can run.
+
 ## Validation
 
 The baseline backend checks are:

SECURITY.md

@@ -13,12 +13,15 @@ Do not expose MoDiff directly to an untrusted LAN, the public internet, a shared
 MoDiff is designed to execute Python and model code:
 
 - Custom-module installation can clone a Git repository or copy a local directory into `custom/`, then import it into the live registry.
-- Enabling `trust_remote_code`, and custom Modular Diffusers block paths that require remote code, can execute Python supplied by a model repository. MoDiff requires an exact 40-character commit revision for these remote custom paths; do not weaken that check to accept moving branches or tags.
+- Reviewed model-execution libraries maintained by Hugging Face run in the backend process with the same filesystem, network, CPU, and accelerator access as MoDiff. Official maintenance reduces neither package supply-chain risk nor the need to review the selected version and integration.
+- Repository-supplied Python would run with backend-process permissions. Current custom Modular Diffusers paths are `contract_only` and reject `trust_remote_code` before model construction; exact cached 40-character commits are still required for Hub contract preview. Do not weaken that fail-closed boundary or accept moving branches/tags if executable support is added later.
 - Model deserialization and optional native/CUDA packages have their own supply-chain and memory-safety risks.
 - Workflows can allocate substantial CPU, RAM, accelerator memory, disk, and network bandwidth.
 
 Install only sources you trust. Review repository ownership, code, dependencies, model licenses, and the exact revision before installation. Prefer immutable commit revisions over moving branches. Disabling a module after import does not undo code that has already run; restart the backend after changing trusted code.
 
+A package being part of the Hugging Face ecosystem is distinct from a model being hosted on the Hub. Do not treat a Hub namespace, model card, or `trust_remote_code` implementation as first-party library code. Optional model runtimes, including Transformers, require an explicit local install action and version verification; template browsing, registry discovery, and Auto planning must remain non-installing operations.
+
 ## Tokens and secrets
 
 The Models UI can validate a Hugging Face token and writes it to `[huggingface] token` in `config.ini`. That file is ignored by Git, but the token is plaintext and is not protected by an operating-system credential store.

data/graphs/studio/ace-step-audio-pipeline/audio-continuation.json

@@ -540,7 +540,7 @@
  "display": "slider",
  "label": "Shift",
  "max": 10,
- "min": 0,
+ "min": 0.1,
  "step": 0.1,
  "type": "float",
  "value": 3

data/graphs/studio/ace-step-audio-pipeline/audio-repaint.json

@@ -452,7 +452,7 @@
  "display": "slider",
  "label": "Shift",
  "max": 10,
- "min": 0,
+ "min": 0.1,
  "step": 0.1,
  "type": "float",
  "value": 3

data/graphs/studio/ace-step-audio-pipeline/audio-variation.json

@@ -452,7 +452,7 @@
  "display": "slider",
  "label": "Shift",
  "max": 10,
- "min": 0,
+ "min": 0.1,
  "step": 0.1,
  "type": "float",
  "value": 3

data/graphs/studio/ace-step-audio-pipeline/text-to-audio--ace-step-chinese-new-year-lora.json

@@ -452,7 +452,7 @@
  "display": "slider",
  "label": "Shift",
  "max": 10,
- "min": 0,
+ "min": 0.1,
  "step": 0.1,
  "type": "float",
  "value": 3
@@ -1731,6 +1731,13 @@
  "type": "bool",
  "value": true
  },
+ "revision": {
+ "default": "",
+ "description": "Required immutable 40-character commit SHA for Hub LoRAs.",
+ "label": "Revision",
+ "type": "string",
+ "value": "cb829a12775740c830a6d49795f16913065dc492"
+ },
  "scale": {
  "default": 0.7,
  "display": "slider",

data/graphs/studio/ace-step-audio-pipeline/text-to-audio--ace-step-custom-lora.json

@@ -452,7 +452,7 @@
  "display": "slider",
  "label": "Shift",
  "max": 10,
- "min": 0,
+ "min": 0.1,
  "step": 0.1,
  "type": "float",
  "value": 3

data/graphs/studio/ace-step-audio-pipeline/text-to-audio.json

@@ -430,7 +430,7 @@
  "display": "slider",
  "label": "Shift",
  "max": 10,
- "min": 0,
+ "min": 0.1,
  "step": 0.1,
  "type": "float",
  "value": 3

data/graphs/studio/flux-dev-pipeline/text-to-image--flux-lora-cinematic-octane-3d.json

@@ -1732,7 +1732,7 @@
  },
  "expected_sha256": {
  "default": "",
- "description": "Optional immutable hash for the selected adapter weight file.",
+ "description": "Required SHA-256 for Hub adapter weights; optional for local weights.",
  "label": "Expected SHA-256",
  "type": "string",
  "value": "64e2788c9d236a3e5f62323baa8853116e2a9c7df41cd2a048141eb22cd46d89"
@@ -1757,6 +1757,13 @@
  "type": "boolean",
  "value": false
  },
+ "revision": {
+ "default": "",
+ "description": "Required immutable Hub commit for the selected adapter repository.",
+ "label": "Revision",
+ "type": "string",
+ "value": "6b32a1624d3fdfab4d518223e8311731dd432cd8"
+ },
  "scale": {
  "default": 1,
  "display": "slider",
@@ -1831,7 +1838,7 @@
  },
  "expected_sha256": {
  "default": "",
- "description": "Optional immutable hash for the selected adapter weight file.",
+ "description": "Required SHA-256 for Hub adapter weights; optional for local weights.",
  "label": "Expected SHA-256",
  "type": "string",
  "value": "cad317378978ba03438c9f00a4fa5ef0628c4a65937c69b8420feaee5e780f81"
@@ -1856,6 +1863,13 @@
  "type": "boolean",
  "value": true
  },
+ "revision": {
+ "default": "",
+ "description": "Required immutable Hub commit for the selected adapter repository.",
+ "label": "Revision",
+ "type": "string",
+ "value": "3ab70503ba8df37565d0212e2876ec48e35e7cb4"
+ },
  "scale": {
  "default": 1,
  "display": "slider",

data/graphs/studio/flux-dev-pipeline/text-to-image--flux-lora-film-noir.json

@@ -1710,7 +1710,7 @@
  },
  "expected_sha256": {
  "default": "",
- "description": "Optional immutable hash for the selected adapter weight file.",
+ "description": "Required SHA-256 for Hub adapter weights; optional for local weights.",
  "label": "Expected SHA-256",
  "type": "string",
  "value": "2970393ce5376e982a594808c1ff0a87f9cec5ddc0da2c094d2a4305d4079324"
@@ -1735,6 +1735,13 @@
  "type": "boolean",
  "value": true
  },
+ "revision": {
+ "default": "",
+ "description": "Required immutable Hub commit for the selected adapter repository.",
+ "label": "Revision",
+ "type": "string",
+ "value": "7a7ff13bbae807a2db6c2e4918f3e13bb1265e60"
+ },
  "scale": {
  "default": 1,
  "display": "slider",

data/graphs/studio/flux-dev-pipeline/text-to-image--flux-lora-ghibli-story.json

@@ -1710,7 +1710,7 @@
  },
  "expected_sha256": {
  "default": "",
- "description": "Optional immutable hash for the selected adapter weight file.",
+ "description": "Required SHA-256 for Hub adapter weights; optional for local weights.",
  "label": "Expected SHA-256",
  "type": "string",
  "value": "5216bd7eeb12bf6f18cd5d40cb090831796b28aca7446577d08c5a7e4a09dc63"
@@ -1735,6 +1735,13 @@
  "type": "boolean",
  "value": true
  },
+ "revision": {
+ "default": "",
+ "description": "Required immutable Hub commit for the selected adapter repository.",
+ "label": "Revision",
+ "type": "string",
+ "value": "ed846114c71efc525e7f5a51e274dc976bb970a8"
+ },
  "scale": {
  "default": 1,
  "display": "slider",

data/graphs/studio/flux-dev-pipeline/text-to-image--flux-lora-oil-painting.json

@@ -1710,7 +1710,7 @@
  },
  "expected_sha256": {
  "default": "",
- "description": "Optional immutable hash for the selected adapter weight file.",
+ "description": "Required SHA-256 for Hub adapter weights; optional for local weights.",
  "label": "Expected SHA-256",
  "type": "string",
  "value": "6de4e6d451ad7690db7185cf84235bf9c15c80aaeb69793fb6647fab62bdd704"
@@ -1735,6 +1735,13 @@
  "type": "boolean",
  "value": true
  },
+ "revision": {
+ "default": "",
+ "description": "Required immutable Hub commit for the selected adapter repository.",
+ "label": "Revision",
+ "type": "string",
+ "value": "1118ed195c7304ffc52a6cd42b41a520ef749cd9"
+ },
  "scale": {
  "default": 1,
  "display": "slider",

data/graphs/studio/flux-dev-pipeline/text-to-image--flux-lora-paper-cutout.json

@@ -1710,7 +1710,7 @@
  },
  "expected_sha256": {
  "default": "",
- "description": "Optional immutable hash for the selected adapter weight file.",
+ "description": "Required SHA-256 for Hub adapter weights; optional for local weights.",
  "label": "Expected SHA-256",
  "type": "string",
  "value": "1863f382199698b8b756d98ecd8060698d5928ed30009cd16f0531f142e4057b"
@@ -1735,6 +1735,13 @@
  "type": "boolean",
  "value": true
  },
+ "revision": {
+ "default": "",
+ "description": "Required immutable Hub commit for the selected adapter repository.",
+ "label": "Revision",
+ "type": "string",
+ "value": "5cdd7ac47ad1b99f705ac2a03a39d480d34abb5d"
+ },
  "scale": {
  "default": 1,
  "display": "slider",

data/graphs/studio/flux-dev-pipeline/text-to-image--flux-lora-photoreal-documentary.json

@@ -1710,7 +1710,7 @@
  },
  "expected_sha256": {
  "default": "",
- "description": "Optional immutable hash for the selected adapter weight file.",
+ "description": "Required SHA-256 for Hub adapter weights; optional for local weights.",
  "label": "Expected SHA-256",
  "type": "string",
  "value": "0a83a924b822b70b5e458d27935ebfa7713edaee04ff9f194209525354031eca"
@@ -1735,6 +1735,13 @@
  "type": "boolean",
  "value": true
  },
+ "revision": {
+ "default": "",
+ "description": "Required immutable Hub commit for the selected adapter repository.",
+ "label": "Revision",
+ "type": "string",
+ "value": "1965e17d2e745fcbf8f4004bdbdf603421ef37a8"
+ },
  "scale": {
  "default": 1,
  "display": "slider",

data/graphs/studio/flux-dev-pipeline/text-to-image--flux-lora-retro-comic.json

@@ -1710,7 +1710,7 @@
  },
  "expected_sha256": {
  "default": "",
- "description": "Optional immutable hash for the selected adapter weight file.",
+ "description": "Required SHA-256 for Hub adapter weights; optional for local weights.",
  "label": "Expected SHA-256",
  "type": "string",
  "value": "af31beee9ea67955d36425f25624d2585fa31271680013e5c9731690aeb78f9d"
@@ -1735,6 +1735,13 @@
  "type": "boolean",
  "value": true
  },
+ "revision": {
+ "default": "",
+ "description": "Required immutable Hub commit for the selected adapter repository.",
+ "label": "Revision",
+ "type": "string",
+ "value": "46f73222df6c97c9f56c3bef42a11979ba5d5aeb"
+ },
  "scale": {
  "default": 1,
  "display": "slider",

data/graphs/studio/flux-dev-pipeline/text-to-image--flux-lora-watercolor.json

@@ -1710,7 +1710,7 @@
  },
  "expected_sha256": {
  "default": "",
- "description": "Optional immutable hash for the selected adapter weight file.",
+ "description": "Required SHA-256 for Hub adapter weights; optional for local weights.",
  "label": "Expected SHA-256",
  "type": "string",
  "value": "e63e44417df35456f425329ad4334143a8bc9fd10316345730ade434106b050e"
@@ -1735,6 +1735,13 @@
  "type": "boolean",
  "value": true
  },
+ "revision": {
+ "default": "",
+ "description": "Required immutable Hub commit for the selected adapter repository.",
+ "label": "Revision",
+ "type": "string",
+ "value": "a565b2140a05f1eece244514f90d2e29b3b1d45d"
+ },
  "scale": {
  "default": 1,
  "display": "slider",

data/graphs/studio/qwen-image-edit-plus-modular-pipeline/edit-image.json

@@ -674,7 +674,7 @@
  "params": {
  "expected_sha256": {
  "default": "",
- "description": "Optional immutable hash for the selected adapter weight file.",
+ "description": "Required for Hub weights; local weights are hashed when this descriptor is created.",
  "label": "Expected SHA-256",
  "type": "string",
  "value": "22226e8d05d354bb356627d428809f5afd7819399b077238a2b70a82883a904f"
@@ -688,18 +688,6 @@
  "model": {
  "display": "modelselect",
  "fieldOptions": {
- "filter": {
- "hub": {
- "className": [
- ""
- ]
- },
- "local": {
- "className": [
- ""
- ]
- }
- },
  "noValidation": true,
  "sources": [
  "hub",
@@ -713,6 +701,13 @@
  "value": "lightx2v/Qwen-Image-Edit-2511-Lightning"
  }
  },
+ "revision": {
+ "default": "",
+ "description": "Required exact commit for a Hub LoRA; unused for a local Safetensors file.",
+ "label": "Revision",
+ "type": "string",
+ "value": "d74eba145674fd7e31b949324e148e21e7118abd"
+ },
  "scale": {
  "default": 1,
  "display": "slider",