Heygen

Heygen

Agent-callable tools for HeyGen's AI video platform, over the public HeyGen v3 API (https://api.heygen.com). Generate AI avatar videos from a script or audio, generate short cinematic clips from a prompt, translate and lip-sync existing videos, drive the prompt-to-video Video Agent, synthesize speech, and browse the avatars and voices those jobs need. Video generation is asynchronous: a create tool returns an id, and you poll the matching get* tool until the job is completed and its result URLs appear.

Independent, unofficial connector for Heygen. Not affiliated with, endorsed by, or sponsored by Heygen. "Heygen" is a trademark of its owner, used only to identify the service this connector works with.

When to use this

• Create AI videos — an avatar (or animated image) speaking a script (createVideo), a short prompt-driven cinematic clip (createCinematicVideo), or a fully agent-authored video from a text prompt (createVideoAgentVideo).
• Transform existing videos — translate into other languages (translateVideo) or replace the audio and re-sync the lips (createLipsync).
• Synthesize speech and pick voices/avatars — text-to-speech (generateSpeech), and browse/clone/design voices and browse avatar looks to resolve the ids the create tools need.
• Track async jobs and account state — poll a video/translation/lipsync/session by id, list past jobs, and check remaining credits (getCurrentUser) before generating.

Setup

This is an agentskills.io skill.

If the connector has not been installed as a skill yet, install it first with npx skills add zapier/connectors --skill heygen (or your harness's own skill-install mechanism), then continue here. Installing the skill copies these files, not dependencies. Before running the CLI, a local MCP server, or zapier-sdk auth commands, run npm install --omit=dev here once. Importing the published package as a dependency in your own project instead? That npm install already resolves everything — see references/use-as-sdk.md.

Want the actual repo source instead — to browse references/, run this connector's tests, or hack on it? See README.md for a scoped git clone.

The connector runs on Node.js 22.18+. Pick the reference that matches how you're running it, and load it before doing anything else:

Scripts

All scripts use the single heygen connection. Generate tools are asynchronous — pair each with its poll tool (e.g. createVideo → getVideo).

Disambiguation & refusals

• Resolve ids before generating; act on a single match. The create tools take ids, not names — get an avatar_id from listAvatarLooks, a voice_id from listVoices, a video_id from listVideos. If a described avatar/voice/video has exactly one match, use it; if two or more tie on the described name/title, list the candidates with a distinguishing field (avatar type, language, created time) and ask which — don't silently pick.
• Don't fake unsupported jobs. This connector does not support: multi-scene "Studio" videos (per-scene avatars/backgrounds in one video), template-based video generation, uploading local asset files (pass a public HTTPS URL to *_url inputs instead), or photo-avatar/digital-twin training pipelines beyond createAvatar's entry point. If asked for one of these, say it isn't supported — don't substitute a different tool and report success for something you didn't do.
• Generation is async and consumes credits. A create call returning an id is not a finished video — poll the get* tool until completed. Failed or in-progress jobs can still consume credits; getCurrentUser reports the remaining balance.

Auth

Every shape passes auth as one connection selector, not the secret — a [<resolver>:]<value> string. Every connector accepts zapier:<connection-id> (Zapier-managed auth — routes through Zapier's auth, retries, and governance layer); some also accept one or more direct-token resolvers (naming and count vary per connector) — check this connector's own resolvers rather than assuming. The <resolver>: prefix is optional; a bare value goes to the first resolver that claims it — a UUID-shaped bare value always claims zapier:. Each script declares the connections it needs and the resolvers each accepts. The exact syntax for passing a connection (and how to see this connector's resolver list) differs by shape — see the reference you loaded above.

Checking what's already configured first? Don't dump environment values to do it — env or env | grep <name> prints the value along with the name, leaking a live credential into the transcript if one is set. Check names only (env | cut -d= -f1 | grep -i <name>) or test a known name directly ([ -n "$VAR_NAME" ]).

Two connection resolvers use different credentials and bill differently: env:HEYGEN_API_KEY (direct — a long-lived API key from the HeyGen dashboard, sent as X-Api-Key, bills the API tier) vs zapier:<connection-id> (Zapier-managed — routes through the connected account's OAuth credential, bills the account's web subscription plan). Both paths are verified. Read the balance field matching your auth path (wallet vs subscription — see references/heygen-api-gotchas.md) before a generate call, to avoid an insufficient_credit failure mid-flow.

No connection yet? Pick one — and follow the reference's own flow to obtain it; never just ask the user for a connection id or token as if they already have one memorized:

Output format

Every script returns a { data, meta } envelope:

• data — the script's result (the shape its outputSchema declares; see the reference you loaded above for how to inspect a script's exact schema in your shape).
• meta.outputDataValidation — what validating data did:
  • { skipped: false, droppedPaths: null } — validated, nothing removed.
  • { skipped: false, droppedPaths: [...], instruction } — validated, but those paths were stripped from data: fields the script returned from the API that the outputSchema doesn't declare. If you need them, re-run with output validation skipped.
  • { skipped: true } — validation was bypassed; data is the raw, unchecked script output.

Reading dropped fields / skipOutputDataValidation. To receive the raw, unvalidated result, opt out of output validation (the exact syntax differs by shape — see the reference you loaded above). Input validation is never skipped.

Trimming the result / filterOutputData. To shrink a large result down to the fields you need, pass a jq expression that post-processes data (again, exact syntax per shape). The jq runs against data only, NOT the { data, meta } envelope, so write it rooted at data (run the script's --help — or your shape's equivalent — to see its output schema). The transformed value replaces data, meta is preserved, and the result is NOT re-validated against the output schema.

References