AgentSkyDocs
OverviewQuickstartCLI referenceAPI referenceAPI GatewayCapabilitiesagent.tomlChannels

Capabilities

What your agent can do out of the box

Built-in capabilities are actions the agent invokes on its own while working — search the web, fetch pages, generate images and video, transcribe voice. You choose which ones an agent gets, and it can run only those: an agent with none granted has none. No API keys or extra subscriptions.

Each action below is listed with its input parameters, what it returns, and its per-call limits. Some actions have more than one provider — the agent uses the default unless asked otherwise. The id on each card is what you put in capabilities = […] in agent.toml, the API's capabilities field, or sky agent create --capability <id> to grant it; the list is empty unless you add ids, and the platform refuses any call outside it. Media outputs are returned as hosted assets the agent can deliver on any channel. Usage is metered per the listed unit and billed from your credits; actions marked with an approval threshold ask before spending more than it.

Web

web.fetchexadefaultread-only · per pageexa.contents

Fetch web page

Retrieve readable public URL content as markdown, text, or JSON metadata.

Parameters

urlurlrequired

Public URL to fetch.

formatenum

Returned content format. — markdown · text · json — default: "markdown"

Returns

urlurl

Fetched URL.

contentstring

Normalized content.

providerstring

Resolved provider.

timeout 30s · up to 2 MB

web.fetchtinyfishread-only · per pagetinyfish.fetch

Fetch JS-heavy page

Explicit JS-heavy/browser fallback for pages Exa cannot fetch statically.

Parameters

urlurlrequired

Public URL to fetch.

formatenum

Returned content format. — markdown · text · json — default: "markdown"

Returns

urlurl

Fetched URL.

contentstring

Normalized content.

providerstring

Resolved provider.

timeout 60s · up to 2 MB

web.browsertinyfishread-only · per minutetinyfish.browser

Browser session

Use only when a page cannot be fetched statically.

Parameters

urlurlrequired

Public URL to open.

taskstring

Browser task to perform. — default: "extract readable content"

Returns

urlurl

Opened URL.

contentstring

Extracted content or result.

providerstring

Resolved provider.

timeout 120s · up to 120s duration

SEO / SERP

Images

image.generategptimagegenerates media · per assetgptimage.generate

Generate image

Create an image from a text prompt through GPT Image.

Parameters

promptstringrequired

Image prompt.

sizeenum

Output image size. — 1024x1024 · 1024x1536 · 1536x1024 — default: "1024x1024"

Returns

assetIdasset-ref

Guild asset id.

mimeTypestring

Output MIME type.

providerstring

Resolved provider.

timeout 120s · up to 20 MB · capped at $1/call · asks for approval above $1 per call

image.remove_backgroundrembgtransforms input · per assetrembg.remove-background

Remove image background

Deterministic background removal for images.

Parameters

imagefilerequired

Input image file.

Returns

assetIdasset-ref

Guild asset id.

mimeTypestring

Output MIME type.

providerstring

Resolved provider.

timeout 60s · up to 20 MB

Video

video.generateseedancegenerates media · per secondseedance.generate

Generate video

Generate a video from a text prompt.

Parameters

promptstringrequired

Video prompt.

durationinteger

Duration in seconds. — default: 5 — 1–10

ratioenum

Aspect ratio. — 16:9 · 9:16 · 1:1 — default: "16:9"

Returns

assetIdasset-ref

Guild asset id.

mimeTypestring

Output MIME type.

providerstring

Resolved provider.

timeout 300s · up to 10s duration · capped at $5/call · asks for approval above $5 per call

video.from_imagemmgenerates media · per secondmm.i2v

Image to video

Generate a video from a still image and a motion prompt.

Parameters

imagefilerequired

Input image file.

promptstringrequired

Motion/video prompt.

durationinteger

Duration in seconds. — default: 5 — 1–10

Returns

assetIdasset-ref

Guild asset id.

mimeTypestring

Output MIME type.

providerstring

Resolved provider.

timeout 300s · up to 10s duration · capped at $5/call · asks for approval above $5 per call

Voice

voice.transcribefish-audiotransforms input · per minutefish-audio.transcribe

Transcribe voice with Fish Audio

Default transcription for audio and video files.

Parameters

audiofilerequired

Audio or video file to transcribe.

languagestring

Language code or auto. — default: "auto"

formatenum

Transcript output format. — text · json · srt — default: "text"

Returns

textstring

Transcript text.

languagestring

Detected or requested language.

providerstring

Resolved provider.

timeout 120s · up to 50 MB · up to 3600s duration