What makes Phosor AI notable?

1.2K downloads. Trust evidence available

How should Phosor AI be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

agentCLAWHUBUnverified

Phosor AI

Generate AI videos, images and speech (text-to-video, image-to-video, reference-to-video, speech-to-video, animate, text-to-image, image-to-image, image edit, text-to-speech), bring your own LoRA models, and generate AI product/model photography for e-commerce (Image Studio) via the Phosor AI platform. Use when the user wants to create videos or images from text prompts, animate images, generate lip-synced video from audio, synthesize speech from text, generate images with a custom LoRA, generate product photography or model/clothing photography for e-commerce listings, or manage generation jobs.

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

1.2k

Updated

Oct 10, 2026

Version

1.3.2

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 1.2K downloads reported by the source. Last updated 10/10/2026.

Avoid when

Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor: Clawhubvendor · observed Oct 10, 2026
Protocol compatibility: OpenClawcompatibility · observed Oct 10, 2026
Adoption signal: 1.2K downloadsadoption · observed Oct 10, 2026
Latest release: 1.3.2release · observed Sep 26, 2026
Handshake status: UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s179vhg33k8dh6daa97nftg9kn8dpdg5:phosor-ai-skills

Install using `clawhub skill install s179vhg33k8dh6daa97nftg9kn8dpdg5:phosor-ai-skills` in an isolated environment before connecting it to live workloads.
No published capability contract is available yet, so validate auth and request/response behavior manually.
Review the upstream CLAWHUB listing at https://clawhub.ai/phosor.ai/phosor-ai-skills before using production credentials.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-phosor-ai-phosor-ai-skills/snapshot"

Documentation

CLAWHUB

145,707 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: phosor-ai-skills
description: Generate AI videos, images and speech (text-to-video, image-to-video, reference-to-video, speech-to-video, animate, text-to-image, image-to-image, image edit, text-to-speech), bring your own LoRA models, and generate AI product/model photography for e-commerce (Image Studio) via the Phosor AI platform. Use when the user wants to create videos or images from text prompts, animate images, generate lip-synced video from audio, synthesize speech from text, generate images with a custom LoRA, generate product photography or model/clothing photography for e-commerce listings, or manage generation jobs.
license: MIT-0
compatibility: Requires Python 3.7+ and network access to phosor.ai
metadata:
  author: phosor.ai
  version: "1.3.2"
  api_version: "v1.2.2"
  homepage: https://phosor.ai
---

# Phosor AI

Generate AI videos and images (text-to-video, image-to-video, speech-to-video, animate, text-to-image, image-to-image), bring your own LoRA models, and generate AI product/model photography for e-commerce (Image Studio) via the Phosor AI platform.

For detailed API endpoints, parameters, pricing, and limits, see [references/api.md](references/api.md).

## Setup

Set your API key:

```bash
export PHOSOR_API_KEY="your-api-key-here"
```

Get an API key at [phosor.ai](https://phosor.ai) → Settings → API Keys.

The CLI script is at `scripts/phosor_client.py`. All commands output JSON to stdout.

## Base URL

The client talks to `https://phosor.ai` by default, over HTTPS. Get your key from phosor.ai → Settings → API Keys.

```bash
python3 scripts/phosor_client.py --api-key <your-key> check-key
```

To point the same client at another Phosor endpoint, pass `--base-url <url>` (or set `PHOSOR_BASE_URL`).
HTTPS is required; plain `http://` is accepted only for `localhost`, or when you explicitly add `--allow-http`.

Note: `studio-analyze` / `studio-suite` `--image-url` must be the **full https S3 URL** returned by `upload-image`, not the bare S3 key path — a bare key errors with `unsupported URL scheme`.

## Quick Start

### MiniMax H3 — Text-to-Video

H3 is **duration-based**, not frame-based: it ignores `--num-frames` / `--fps` (output is
always 24fps) and bills per output second. Pick the frame size with
`--resolution-tier` + `--aspect` instead of `--width/--height`.

```bash
python3 scripts/phosor_client.py submit "A cat walking on a beach at sunset" \
  --model minimax/h3/text-to-video \
  --resolution-tier 768p --aspect 16:9 --duration 5
```

### MiniMax H3 — Image-to-Video

```bash
# Upload first (direct URLs are not accepted); then submit with the returned s3_key
python3 scripts/phosor_client.py upload-image /path/to/first-frame.jpg

python3 scripts/phosor_client.py submit "The person starts dancing" \
  --model minimax/h3/image-to-video \
  --image-url "images/img-xxx.jpg" \
  --end-image-url "images/img-yyy.jpg" \
  --resolution-tier 480p --aspect 9:16 --duration 6
```

`--end-image-url` is optional and pins the cl

README.md

# Phosor AI — Agent Skill

**Skill version 1.3.2** · API v1.2.2 · updated 2026-09-25

Two version numbers, on purpose: the skill version tracks this package
(commands, docs, the bundled client), the API version tracks the gateway
contract. A skill release that only rewords docs or fixes the client does not
move the API version, and a gateway change does not force a skill release.

Verify what you installed: `python3 scripts/phosor_client.py --version`

## Quick Start

```bash
export PHOSOR_API_KEY="your-key"

# Text-to-Video
python3 scripts/phosor_client.py submit "A cat walking on a beach" --width 854 --height 480

# Image-to-Video (two-step: upload then submit)
python3 scripts/phosor_client.py upload-image photo.jpg
python3 scripts/phosor_client.py submit "The scene comes alive" --image-url "images/img-xxx.jpg"

# Check status / get result
python3 scripts/phosor_client.py status <request_id>
python3 scripts/phosor_client.py result <request_id>

# LoRA Upload (custom pre-trained)
python3 scripts/phosor_client.py upload-lora high_noise.safetensors low_noise.safetensors

python3 scripts/phosor_client.py save-lora <lora_id>

# Image Studio: AI product photography (separate product surface, same API key)
python3 scripts/phosor_client.py upload-image product.jpg
python3 scripts/phosor_client.py studio-suite --image-url "<s3_url from upload>" \
  --layout-types "white_background,lifestyle_scene" --count-per-type 2
python3 scripts/phosor_client.py studio-status <job_id>
```

See [SKILL.md](SKILL.md#image-studio-product--model-photography) for the full Image Studio quick start (clothing/model suite, one-off edits like remove-bg/inpaint/translate).

## Requirements

- Python 3.7+ (stdlib only, no pip install needed)
- `PHOSOR_API_KEY` environment variable

## Commands

Run `python3 scripts/phosor_client.py --help` for all 31 commands (23 video/LoRA + 8 Image Studio).

## Links

- [Phosor AI](https://phosor.ai)
- [API Documentation](https://docs.phosor.ai)

_meta.json

{
  "ownerId": "kn7dkx5wmnapr9y2x91m6qnb8s82w04e",
  "slug": "phosor-ai-skills",
  "version": "1.3.2",
  "publishedAt": 1790399871799
}

references/api.md

# Phosor AI API Reference

Base URL: `https://phosor.ai`

All endpoints require `X-API-Key` header unless noted otherwise.

## Endpoints

### Models

| Method | Path | Auth | Description |
|--------|------|------|-------------|
| GET | `/api/v1/models` | None | List available models |

### Inference

| Method | Path | Description |
|--------|------|-------------|
| POST | `/api/v1/inference/submit` | Submit video or image generation job |
| GET | `/api/v1/inference/status/{request_id}` | Get job status + progress |
| GET | `/api/v1/inference/result/{request_id}` | Get completed result (video or image URL) |
| GET | `/api/v1/inference/history` | Get user's job history |

### Storage — Image / Audio

| Method | Path | Description |
|--------|------|-------------|
| POST | `/api/v1/storage/image/upload` | Multipart image or audio upload (images: jpg/png/webp; audio for S2V: mp3/wav/flac/aac/ogg/m4a) |
| POST | `/api/v1/storage/image/import` | Import from public URL |

### Storage — LoRA

| Method | Path | Description |
|--------|------|-------------|
| POST | `/api/v1/storage/lora/upload` | Upload two .safetensors files (video LoRA: high_noise + low_noise) |
| POST | `/api/v1/storage/lora/import` | Import from HTTPS URLs (video: two files, image: single file) |

### LoRA Management

| Method | Path | Description |
|--------|------|-------------|
| GET | `/api/v1/loras` | List LoRA models |
| GET | `/api/v1/loras/{lora_id}` | Get single LoRA details |
| GET | `/api/v1/loras/{lora_id}/status` | Get processing status |
| POST | `/api/v1/loras/{lora_id}/save` | Activate a LoRA (extend expiry to 7 days) |
| DELETE | `/api/v1/loras/{lora_id}` | Soft delete |

### Image Studio (product & model photography — separate product surface)

All paths below are relative to `/api/v1/image-studio` (e.g. the full path for `/product/suite` is `/api/v1/image-studio/product/suite`). **Every path requires `X-API-Key`, including `GET /pricing`** — there is no unauthenticated Image Studio endpoint.

All generation/analyze endpoints are async: POST returns `{"request_id": "...", "status": "pending"}` immediately; poll `GET /jobs/{request_id}` until `status` is `"done"`, `"error"` or `"cancelled"`.

> **The async key is `request_id`, not `request_id`.** Earlier revisions of this document said
> `request_id`; that key is never present in a response. Reading it yields `undefined`, and the
> poll then never resolves. Image Studio's `request_id` is a separate namespace from the
> video/LoRA `request_id` — do not pass one to the other's status endpoint.

| Method | Path | Description |
|--------|------|-------------|
| POST | `/agent/analyze` | AI-analyze an image for the Agent-image workflow (freemium) |
| POST | `/product/analyze` | AI-analyze a product image ahead of `product/suite` (freemium) |
| POST | `/model/analyze` | AI-analyze a garment image ahead of `model/clothing-suite` (freemium) |
| POST | `/product/reference/analyze` | Analyze a reference product page/image for s

CHANGELOG.md

# Changelog

Skill package versions. The API contract version is separate and lives in
`SKILL.md` as `metadata.api_version` — a skill release does not move it.

## 1.3.2 — 2026-09-25 (API v1.2.2)

`minimax/h3/reference-to-video/turbo` removed

- Requests to `minimax/h3/reference-to-video/turbo` now return an unknown-model error
- Use `minimax/h3/reference-to-video` instead — same inputs and pricing structure
- `reference-to-video` (non-turbo) is unaffected
- API contract version v1.2.2: a model id was removed; no other endpoints or parameters changed

## 1.3.1 — 2026-09-24 (API v1.2.1)

Reference-to-Video limits and pricing

- Reference caps, both 480p and 768p: 9 images when sending only images, 5 images when
  videos or audio are also present, 3 reference videos, 3 reference audios
- Total reference video length is 15.1s across all reference videos
- `reference-to-video` now uses the same rates as the rest of H3: 480p $0.0045/s,
  768p $0.012/s, reference images $0.0075 each after the first 5
- Bundled client validation and version string updated to match
- API contract version v1.2.1: Reference-to-Video limits and pricing changed; no new endpoints or parameters

## 1.3.0 — 2026-09-22 (API v1.2.0)

Reference metadata and H3 limits

- Reference labels are forwarded in the same order as image, video, and audio URLs
- 768p Ref2VA supports the same 9 image-only, 4 mixed-image, and 3 video caps as 480p
- Correct the bundled client's stale 768p validation and 15-second video budget

## 1.2.0 — 2026-09-13 (API v1.2.0)

Reference-to-Video

- Total reference video length raised from 10s to 15s, across all reference
  videos combined. `reference-to-video/turbo` is unchanged at 10s.
- No other caps moved, and pricing is unchanged.

## 1.2.0 — 2026-09-10

Pricing

- MiniMax H3 (T2V / I2V / Ref2VA / Ref2VA Turbo) now shares one per-output-second
  rate regardless of clip length: 480p $0.0045/s, 768p $0.012/s (previously 0.02 / 0.04 USD
  for T2V/I2V and 0.025 / 0.063 USD for Ref2VA)
- Ref2VA reference inputs: the first 5 reference images are free, then $0.0075 each
  (previously 0.01 USD from the first image); reference audio is free (previously 0.01 USD);
  reference video stays at the output tier rate
- `GET /api/v1/pricing/config` adds `reference_image_free_count`; the built-in
  client's price table and estimate formula follow it

Reference-to-Video Turbo

- `minimax/h3/reference-to-video/turbo` added to the model table with its own reference
  caps and 768p usage guidance (stay at or below 3 images, or 2 images plus 1 video)

Client and docs

- Environment section and `--base-url` help use placeholders instead of a specific host
- `studio-layouts` on a self-hosted deployment reads `PHOSOR_WEB_BASE_URL` for the web
  front instead of guessing ports
- Code comments in `phosor_client.py` are English throughout

## 1.1.0 — 2026-09-03

New models

- MiniMax H3: text-to-video, image-to-video with an optional closing frame,
  and Reference-to-Video, which tak

Github ReposUpdated 1d agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW

Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW

Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.