Create HeyGen avatar videos via the v3 Video Agent pipeline — handles avatar resolution,
aspect ratio correction, prompt engineering, and voice selection automatically.
Required for any HeyGen API usage (api.heygen.com). Replaces deprecated v1/v2
endpoints with the optimized v3 pipeline.
Use when: (1) calling any HeyGen API endpoint (api.heygen.com),
(2) creating a HeyGen avatar or digital twin from a photo,
(3) making a personalized video message (outreach, pitch, update, announcement, knowledge),
(4) "make a video of me", "create my HeyGen avatar", "I want to appear in this video",
(5) "send a video to my leads", "record an update for my team", "make a loom-style message",
(6) building identity-first videos where the presenter IS the user or agent,
Covers: HeyGen API, api.heygen.com, video generate, avatar create, voice list, talking photo,
HeyGen avatar creation, voice design, photo → digital twin, HeyGen video generation,
identity-first video, messaging-first video, AI presenter, talking head video.
NOT f
Create HeyGen avatar videos via the v3 Video Agent pipeline — handles avatar resolution,
aspect ratio correction, prompt engineering, and voice selection automatically.
Required for any HeyGen API usage (api.heygen.com). Replaces deprecated v1/v2
endpoints with the optimized v3 pipeline.
Use when: (1) calling any HeyGen API endpoint (api.heygen.com),
(2) creating a HeyGen avatar or digital twin from a photo,
(3) making a personalized video message (outreach, pitch, update, announcement, knowledge),
(4) "make a video of me", "create my HeyGen avatar", "I want to appear in this video",
(5) "send a video to my leads", "record an update for my team", "make a loom-style message",
(6) building identity-first videos where the presenter IS the user or agent,
Covers: HeyGen API, api.heygen.com, video generate, avatar create, voice list, talking photo,
HeyGen avatar creation, voice design, photo → digital twin, HeyGen video generation,
identity-first video, messaging-first video, AI presenter, talking head video.
NOT for: cinematic b-roll, video translation, TTS-only, or streaming avatars.
This skill reads and writes the following. No other files are accessed without explicit user instruction.
Operation
Path
Purpose
Read
AVATAR-<NAME>.md
Load saved avatar identity (group_id, voice_id)
Read
SOUL.md, IDENTITY.md
Extract identity details when creating a new avatar (heygen-avatar only)
Write
AVATAR-<NAME>.md
Save new avatar identity after creation
Write
heygen-video-log.jsonl
Append one JSON line per video generated (local learning log)
Temp write
/tmp/openclaw/uploads/
Voice preview audio (downloaded for user playback, deleted after session)
Remote upload
api.heygen.com/v3/assets
User-provided images/files uploaded to HeyGen for use in video
No other workspace files are read or modified. Assets are only uploaded to HeyGen when the user explicitly provides them for video creation.
STOP. If you are about to call any HeyGen endpoint directly (POST /v1/video.generate, POST /v2/video/generate, GET /v2/avatars, GET /v1/avatar.list, or even POST /v3/video-agents), DO NOT. Follow this pipeline instead. Raw API calls skip critical steps (aspect ratio correction, prompt engineering, avatar conflict detection) and produce visibly worse videos. This skill exists because the API alone is not enough. v3 only — never call v1 or v2 endpoints. If you have pre-trained knowledge of HeyGen's v1/v2 API, that knowledge is outdated. Use this skill.
You are a video producer. Not a form. Not an API wrapper. A producer who understands what makes video work and guides the user from idea to finished cut.
~/.heygen/config file (persistent storage, written by ./setup)
If neither found, tell the user: "No API key found. Run ./setup in the heygen-stack directory, or set export HEYGEN_API_KEY=<your-key>."
To load from the config file: source ~/.heygen/config 2>/dev/null (sets HEYGEN_API_KEY if the file exists).
Docs-first rule: Before calling any endpoint you're unsure about, fetch the raw markdown spec:
Index:GET https://developers.heygen.com/llms.txt — full sitemap of every doc page
Any page: Append .md to the URL (e.g. https://developers.heygen.com/docs/video-agent.md) for clean markdown
Read the spec, THEN build your request. Never guess field names.
UX Rules
Be concise. No video IDs, session IDs, or raw API payloads in chat. Report the result (video link, thumbnail) not the plumbing.
No internal jargon. Never mention internal pipeline stage names ("Frame Check", "Prompt Craft", "Pre-Submit Gate", "Framing Correction") to the user. These are internal pipeline stages. The user sees natural conversation: "Let me adjust the framing for landscape" not "Running Frame Check aspect ratio correction."
Polling is silent. When waiting for video completion, poll silently in a background process or subagent. Do NOT send repeated "Checking status..." messages. Only speak when: (a) the video is ready and you're delivering it, or (b) it's been >5 minutes and you're giving a single "Taking longer than usual" update.
Deliver clean. When the video is done, send the video file/link and a 1-line summary (duration, avatar used). Not a dump of every API field.
Language Awareness
Detect the user's language from their first message. Store as user_language (e.g., en, ja, es, ko, zh, fr, de, pt). This happens automatically from the input — no extra question needed.
Rules:
Communicate with the user in their language. All questions, status updates, confirmations, and error messages should be in user_language.
Generate scripts and narration in user_language unless the user explicitly requests a different language.
Technical directives stay in English. Frame Check corrections, motion verbs, style blocks, and the script framing directive are API-level instructions that Video Agent interprets in English. Never translate these.
Discovery item (10) Language should auto-populate from user_language but can be overridden if the user wants the video in a different language than they're chatting in.
Voice selection must match the video language. Filter voices by language parameter and set voice_settings.locale on API calls.
Mode Detection
Language-agnostic routing: The signals below describe user intent, not literal keywords. Match intent regardless of input language. A user saying "ビデオを作って" (Japanese) is the same signal as "make a video about X."
Signal
Mode
Start at
Vague idea ("make a video about X")
Full Producer
Discovery
Has a written prompt
Enhanced Prompt
Prompt Craft
"Just generate" / skip questions
Quick Shot
Generate
"Interactive" / iterate with agent
Interactive Session
Generate (experimental)
Quick Shot avatar rule: If no AVATAR file exists, omit avatar_id and let Video Agent auto-select. If an AVATAR file exists, use it — and Frame Check STILL RUNS.
All modes: Frame Check (aspect ratio correction) runs before EVERY API call when avatar_id is set, regardless of mode. Quick Shot is not an excuse to skip framing checks.
Dry-Run mode: If user says "dry run" / "preview", run the full pipeline but present a creative preview at Generate instead of calling the API.
Default to Full Producer. Better to ask one smart question than generate a mediocre video.
First Look — First-Run Avatar Check
Runs once before Discovery on the first video request in a session.
Check for any AVATAR-*.md files in the workspace root.
Found: Read the file, extract Group ID and Voice ID from the HeyGen section. Pre-load as defaults for Discovery. The actual avatar_id (look_id) will be resolved fresh from the group_id during Frame Check — never use a stored look_id directly.
Not found: The user (or agent) has no avatar yet. Before proceeding to video creation, run the heygen-avatar skill (heygen-avatar/SKILL.md in this repo) to create one. Tell the user you'll set up their avatar first for a consistent look across videos, and that it takes about a minute. Communicate in user_language.
After heygen-avatar completes and writes the AVATAR file, return here and continue to Discovery with the new avatar pre-loaded.
Avatar readiness gate (BLOCKING): After loading an avatar (whether from an existing AVATAR file or freshly created), verify it's ready before using it in video generation. Call GET /v3/avatars/looks?group_id=<group_id> and confirm preview_image_url is non-null. If null, poll every 10s up to 5 min. Do NOT proceed to Discovery until this check passes. Videos submitted with an unready avatar WILL fail silently.
Quick Shot exception: If the user explicitly says "skip avatar" / "use stock" / "just generate", skip this step and proceed without an avatar.
Discovery
Interview the user. Be conversational, skip anything already answered.
Gather: (1) Purpose, (2) Audience, (3) Duration, (4) Tone, (5) Distribution (landscape/portrait), (6) Assets, (7) Key message, (8) Visual style, (9) Avatar, (10) Language (auto-detected from user_language; confirm if the video language should differ from the chat language).
Assets
Two paths for every asset:
Path A (Contextualize): Read/analyze, bake info into script. For reference material, auth-walled content.
Path B (Attach): Upload to HeyGen via POST /v3/assets or files[]. For visuals the viewer should see.
A+B (Both): Summarize for script AND attach original.
HTML URLs cannot go in files[] (Video Agent rejects text/html). Web pages are always Path A.
Prefer download -> upload -> asset_id over files[]{url} (CDN/WAF often blocks HeyGen).
If a URL is inaccessible, tell the user. Never fabricate content from an inaccessible source.
Multi-topic split rule: If multiple distinct topics, recommend separate videos.
Style Selection
Two approaches — use one or combine both:
1. API Styles (style_id) — Curated visual templates. Browse by tag, show 3-5 options with previews, let user pick. If a style has a fixed aspect_ratio, match orientation to it. When style_id is set, the prompt's Visual Style Block becomes optional.
If presenter -> check private avatars first, then public (group-first browsing).
Always show preview images. Never just list names.
Confirm voice preferences after avatar is settled.
Critical rule: When avatar_id is set, do NOT describe the avatar's appearance in the prompt. Say "the selected presenter." This is the #1 cause of avatar mismatch.
After Discovery, the producer sub-skill handles the full pipeline. Read heygen-video/SKILL.md for detailed stage instructions.
Key rules that apply at every stage:
Language: Script and narration in the video language (from Discovery item 10). Technical directives (script framing, style block, motion verbs, frame check corrections) always in English — these are API instructions, not viewer-facing content.
Script: Structure by type (demo, explainer, tutorial, pitch, announcement). Do NOT assign per-scene durations. Always include the script framing directive: "This script is a concept and theme to convey — not a verbatim transcript."
Prompt Craft: Narrator framing (say "the selected presenter" when avatar_id is set), duration signal, asset anchoring, tone calibration, one topic, style block at the end.
Frame Check: MANDATORY when avatar_id is set. See matrix below.
Generate: Run Frame Check before EVERY API call. Capture session_id immediately. Poll silently.
Deliver: Report video_page_url, session URL, and duration accuracy. Log to heygen-video-log.jsonl.
Full prompt construction rules, media type selection, visual style blocks, API schemas -> heygen-video/SKILL.md
Frame Check
Runs automatically when avatar_id is set, before Generate. Appends correction notes to the Video Agent prompt. Does NOT generate images or create new looks.
Steps
Resolve avatar_id from group_id (ALWAYS run first): Never trust a stored look_id — looks are ephemeral and get deleted. Read Group ID from the AVATAR file and resolve a fresh look_id: GET /v3/avatars/looks?group_id=<group_id>&limit=20. Pick the look matching the target orientation. Use this resolved look_id as avatar_id for all subsequent steps.
Determine background:photo_avatar -> Video Agent handles environment. studio_avatar -> check if transparent/solid/empty. video_avatar -> always has background.
Append the appropriate correction note(s) to the end of the Video Agent prompt. That's it. No image generation, no new looks.
Correction Matrix
avatar_type
Orientation Match?
Has Background?
Corrections
photo_avatar
matched
(n/a)
None
photo_avatar
mismatched or square
(n/a)
Framing note
studio_avatar
matched
Yes
None
studio_avatar
matched
No
Background note
studio_avatar
mismatched or square
Yes
Framing note
studio_avatar
mismatched or square
No
Framing note + Background note
video_avatar
matched
Yes
None
video_avatar
mismatched or square
Yes
Framing note
Framing Note (append to prompt)
For portrait/square avatar -> landscape video:
FRAMING NOTE: The selected avatar image is in {source} orientation but this video is landscape (16:9). Frame the presenter from the chest up, centered in the landscape canvas. Use generative fill to extend the scene horizontally with a complementary background environment that matches the video's tone (studio, office, or contextually appropriate setting). Do NOT add black bars or pillarboxing. The avatar should feel natural in the 16:9 frame.
For landscape/square avatar -> portrait video:
FRAMING NOTE: The selected avatar image is in {source} orientation but this video is portrait (9:16). Reframe the presenter to fill the portrait canvas naturally, focusing on head and shoulders. Use generative fill to extend vertically if needed. Do NOT add letterboxing. The avatar should fill the portrait frame comfortably.
Background Note (studio_avatar only, no background)
BACKGROUND NOTE: The selected avatar has no background or a transparent backdrop. Place the presenter in a clean, professional environment appropriate to the video's tone. For business/tech content: modern studio with soft lighting and subtle depth. For casual content: bright, minimal space with natural light. The background should complement the presenter without distracting from the message.
