Sogni Creative Agent for Hermes
Generate images, video, and music through Sogni AI's decentralized GPU network.
Invoke sogni-agent-hermes for every command in this skill. Fall back to
sogni-agent only when the Hermes launcher is not on PATH. The two commands
have identical flags and output; the Hermes launcher adds host attribution.
Set up the CLI
Only perform setup when the user asks to install or upgrade the skill, or when
a command fails because the CLI is missing.
Install the CLI:
npm install -g @sogni-ai/sogni-creative-agent-skill@latest
sogni-agent-hermes --version
Ask the user to get an API key from https://dashboard.sogni.ai by opening
the account menu. Accept either SOGNI_API_KEY in the environment or the
credentials file ~/.config/sogni/credentials containing
SOGNI_API_KEY=<key>. Never echo, log, commit, or embed the key in generated
files.
Run sogni-agent-hermes doctor --json after installation or upgrade and
confirm that success is true. Do not run doctor as a routine preflight;
normal generation commands perform their own checks.
Execute requests
- Save final media in the user's current working directory unless they request
another destination. Never put final output in
/tmp.
- Use paths the user supplied. If an agent host exposes inbound attachments,
run
sogni-agent-hermes --json --list-media instead of guessing filenames.
- Honor an explicitly requested model. Otherwise use the CLI defaults and
consult models.md only when model selection matters.
- Plan multi-step work in Hermes, then invoke the focused CLI command. Use the
hosted workflow API only when its durable orchestration or replay behavior is
useful.
- Parse
--json output when subsequent steps need exact job IDs, paths, URLs,
costs, or error fields.
- Return the final local path and the important render settings. On failure,
preserve the CLI's error and actionable recovery hint.
Common commands
# Generate or edit an image
sogni-agent-hermes -o ./image.png "cinematic moonlit mountain lake"
sogni-agent-hermes -c ./source.png -o ./edited.png "make it night; add fireflies"
# Promptless RTX VSR upscale
sogni-agent-hermes --upscale ./source.png -o ./upscaled.png
# Generate or animate video
sogni-agent-hermes --video -o ./video.mp4 "a paper dragon takes flight"
sogni-agent-hermes --video --ref ./start.png -o ./animated.mp4 "slow camera push-in"
sogni-agent-hermes --video --ref ./first.png --ref-end ./last.png -o ./transition.mp4 "smooth transformation"
sogni-agent-hermes --video -m wan3 --target-resolution 1080 --duration 8 -o ./wan3.mp4 'a presenter says "Welcome" in a detailed studio'
sogni-agent-hermes --video -m wan3-enhanced --target-resolution 1080 --duration 8 --wan3-ratio 16:9 -o ./wan3-enhanced.mp4 'a presenter says "Welcome" in a detailed studio'
# Generate music
sogni-agent-hermes --music -o ./soundtrack.mp3 "30-second ambient synth theme"
# Inspect state and available media
sogni-agent-hermes --last --json
sogni-agent-hermes --json --list-media
sogni-agent-hermes --help
Load detailed guidance only when needed
- Model choice, quality tiers, Krea identity edits, and video aliases: read
models.md.
- Krea 2 LoRA ids, strength ranges, bipolar sliders, and stacking up to 8 in
one render: read krea2-loras.md.
- Video prompts, reference roles, LTX, WAN 2.2, Wan 3, MiniMax, Seedance, and HappyHorse:
read video-prompting.md.
- Prompt-only image deliverables for SD/SDXL, FLUX.1 Schnell, Chroma, Krea 2, Qwen,
Z-Image, GPT Image, and model-specific editing: read
image-prompting.md.
- Local video cutting, stitching, overlays, subtitles, audio remixing, and
verification: read video-editing.md.
- Saved identities, reusable assets, memory, and personality: read
personas-memory.md.
- Hosted chat, durable workflows, structured contracts, and replay records:
read hosted-api.md.
- Turning an image folder into a deduplicated, music-backed seamless reel:
read loop-maker.md completely before starting.
- Seamless textures or tiles: read
seamless-tiling.md.
- Explicit private/adult video requests: read
private-mature-video.md and follow its
eligibility and prompting rules.
Where a reference says sogni-agent, substitute sogni-agent-hermes.
1---2name: sogni-creative-agent-skill-23description: Generate and edit images, upscale images, create video and music, animate image folders into seamless reels, and manage reusable personas with Sogni AI. Use when a Hermes user asks to draw, render, upscale, edit or restyle an image, animate an image, make a video, create music or lyrics, build a media workflow, apply a saved identity, or refine an earlier Sogni result.4license: MIT5---67# Sogni Creative Agent for Hermes89Generate images, video, and music through Sogni AI's decentralized GPU network.1011Invoke `sogni-agent-hermes` for every command in this skill. Fall back to12`sogni-agent` only when the Hermes launcher is not on `PATH`. The two commands13have identical flags and output; the Hermes launcher adds host attribution.1415## Set up the CLI1617Only perform setup when the user asks to install or upgrade the skill, or when18a command fails because the CLI is missing.19201. Install the CLI:2122 ```bash23 npm install -g @sogni-ai/sogni-creative-agent-skill@latest24 sogni-agent-hermes --version25 ```26272. Ask the user to get an API key from https://dashboard.sogni.ai by opening28 the account menu. Accept either `SOGNI_API_KEY` in the environment or the29 credentials file `~/.config/sogni/credentials` containing30 `SOGNI_API_KEY=<key>`. Never echo, log, commit, or embed the key in generated31 files.323. Run `sogni-agent-hermes doctor --json` after installation or upgrade and33 confirm that `success` is `true`. Do not run `doctor` as a routine preflight;34 normal generation commands perform their own checks.3536## Execute requests3738- Save final media in the user's current working directory unless they request39 another destination. Never put final output in `/tmp`.40- Use paths the user supplied. If an agent host exposes inbound attachments,41 run `sogni-agent-hermes --json --list-media` instead of guessing filenames.42- Honor an explicitly requested model. Otherwise use the CLI defaults and43 consult [models.md](references/models.md) only when model selection matters.44- Plan multi-step work in Hermes, then invoke the focused CLI command. Use the45 hosted workflow API only when its durable orchestration or replay behavior is46 useful.47- Parse `--json` output when subsequent steps need exact job IDs, paths, URLs,48 costs, or error fields.49- Return the final local path and the important render settings. On failure,50 preserve the CLI's error and actionable recovery hint.5152## Common commands5354```bash55# Generate or edit an image56sogni-agent-hermes -o ./image.png "cinematic moonlit mountain lake"57sogni-agent-hermes -c ./source.png -o ./edited.png "make it night; add fireflies"5859# Promptless RTX VSR upscale60sogni-agent-hermes --upscale ./source.png -o ./upscaled.png6162# Generate or animate video63sogni-agent-hermes --video -o ./video.mp4 "a paper dragon takes flight"64sogni-agent-hermes --video --ref ./start.png -o ./animated.mp4 "slow camera push-in"65sogni-agent-hermes --video --ref ./first.png --ref-end ./last.png -o ./transition.mp4 "smooth transformation"66sogni-agent-hermes --video -m wan3 --target-resolution 1080 --duration 8 -o ./wan3.mp4 'a presenter says "Welcome" in a detailed studio'67sogni-agent-hermes --video -m wan3-enhanced --target-resolution 1080 --duration 8 --wan3-ratio 16:9 -o ./wan3-enhanced.mp4 'a presenter says "Welcome" in a detailed studio'6869# Generate music70sogni-agent-hermes --music -o ./soundtrack.mp3 "30-second ambient synth theme"7172# Inspect state and available media73sogni-agent-hermes --last --json74sogni-agent-hermes --json --list-media75sogni-agent-hermes --help76```7778## Load detailed guidance only when needed7980- Model choice, quality tiers, Krea identity edits, and video aliases: read81 [models.md](references/models.md).82- Krea 2 LoRA ids, strength ranges, bipolar sliders, and stacking up to 8 in83 one render: read [krea2-loras.md](references/krea2-loras.md).84- Video prompts, reference roles, LTX, WAN 2.2, Wan 3, MiniMax, Seedance, and HappyHorse:85 read [video-prompting.md](references/video-prompting.md).86- Prompt-only image deliverables for SD/SDXL, FLUX.1 Schnell, Chroma, Krea 2, Qwen,87 Z-Image, GPT Image, and model-specific editing: read88 [image-prompting.md](references/image-prompting.md).89- Local video cutting, stitching, overlays, subtitles, audio remixing, and90 verification: read [video-editing.md](references/video-editing.md).91- Saved identities, reusable assets, memory, and personality: read92 [personas-memory.md](references/personas-memory.md).93- Hosted chat, durable workflows, structured contracts, and replay records:94 read [hosted-api.md](references/hosted-api.md).95- Turning an image folder into a deduplicated, music-backed seamless reel:96 read [loop-maker.md](references/loop-maker.md) completely before starting.97- Seamless textures or tiles: read98 [seamless-tiling.md](references/seamless-tiling.md).99- Explicit private/adult video requests: read100 [private-mature-video.md](references/private-mature-video.md) and follow its101 eligibility and prompting rules.102103Where a reference says `sogni-agent`, substitute `sogni-agent-hermes`.