CtrlK
BlogDocsLog inGet started
Tessl Logo

avatar-video

Create AI avatar videos with precise control over avatars, voices, scripts, scenes, and backgrounds using HeyGen's v2 API. Use when: (1) Choosing a specific avatar and voice for a video, (2) Writing exact scripts for an avatar to speak, (3) Building multi-scene videos with different backgrounds per scene, (4) Creating transparent WebM videos for compositing, (5) Using talking photos as video presenters, (6) Integrating HeyGen avatars with Remotion, (7) Batch video generation with exact specs, (8) Brand-consistent production videos with precise control.

66

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

72%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

Well-structured, actionable overview with excellent progressive disclosure and verified references. The main gap is workflow_clarity: validation lives in Best Practices rather than as an integrated checkpoint with a feedback loop in the generation/polling workflow.

Suggestions

Fold input validation into the Default Workflow as an explicit numbered checkpoint (e.g., 'Validate: confirm avatar_id and voice_id exist via GET /v2/avatars and GET /v2/voices before generating') rather than only listing it under Best Practices.

Add a poll-status feedback loop to step 5 (e.g., 'If status is failed, read the error, fix the request body, and re-run POST /v2/video/generate') so error recovery is part of the workflow.

De-duplicate the reference listings — keep either the Quick Reference table or the categorized Reference Files section, not both — to tighten token efficiency.

Show a minimal complete POST /v2/video/generate request body inline (a single-scene example) so the core generation step is copy-paste ready without opening a reference file.

DimensionReasoningScore

Conciseness

Lean body with tables, bullet lists, and a minimal curl example; no padding of concepts Claude already knows. Minor redundancy: the reference files are listed three times (Default Workflow inline links, Quick Reference table, Reference Files section), which is slightly above the lean anchor 5.

4 / 5

Actionability

Concrete endpoints ('GET /v2/avatars', 'POST /v2/video/generate', 'GET /v2/videos/{video_id}'), an executable curl example, and named MCP tools make guidance mostly executable. The full generate POST body is deferred to a reference rather than shown inline, so it stops short of anchor 5.

4 / 5

Workflow Clarity

A clear 5-step Default Workflow with explicit polling exists, and 'Validate inputs' appears in Best Practices, but validation is a loose bullet rather than an integrated checkpoint with a validate->fix->retry feedback loop; for a batch/generation skill the missing integrated checkpoint caps this at anchor 3.

3 / 5

Progressive Disclosure

Clear overview with a Quick Reference table and categorized Reference Files section; all references are one level deep and every referenced file (avatars.md, voices.md, scripts.md, video-generation.md, video-status.md, backgrounds.md, captions.md, text-overlays.md, templates.md, photo-avatars.md, assets.md, remotion-integration.md, webhooks.md, dimensions.md, quota.md) exists in references/, giving easy navigation matching anchor 5.

5 / 5

Total

16

/

20

Passed

Description

92%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, specific description with explicit 'what' and 'when' guidance and an enumerated trigger list. Trigger-term coverage is good but could add a few synonyms and file extensions.

DimensionReasoningScore

Specificity

Lists multiple concrete actions across 8 enumerated capabilities ('Choosing a specific avatar and voice', 'Writing exact scripts', 'Building multi-scene videos', 'Creating transparent WebM videos', 'Using talking photos', 'Integrating HeyGen avatars with Remotion', 'Batch video generation', 'Brand-consistent production') — comprehensive coverage matching anchor 5.

5 / 5

Completeness

Explicitly answers both 'what' ('Create AI avatar videos with precise control... using HeyGen's v2 API') and 'when' ('Use when: (1)...(8)...') with concrete trigger phrases, matching anchor 5.

5 / 5

Trigger Term Quality

Good natural-term coverage ('avatar', 'voice', 'script', 'multi-scene videos', 'transparent WebM', 'talking photos', 'Remotion', 'batch video generation') but missing some synonyms and file extensions (e.g., '.webm', 'HeyGen' as a when-trigger), placing it between anchors 4 and 5.

4 / 5

Distinctiveness Conflict Risk

Clear niche (HeyGen v2 API avatar videos) with distinct triggers and minimal conflict risk; the body even distinguishes it from a sibling create-video skill.

5 / 5

Total

19

/

20

Passed

Validation

75%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation12 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

12

/

16

Passed

Repository
calesthio/OpenMontage
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.