screen-test
Choosing the video model on evidence — the bake-off protocol and the selection table.
Skill by maniculehq
Chalk Labs
Need help deploying screen-test?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/maniculehq-screen-test
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: testing
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/screen-test
What's inside
- Skill name
screen-test- Description
- Choosing the video model on evidence — the bake-off protocol and the selection table. Read before stages 3 and 5.
- License
- MIT
- Instructions
- 22 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Screen testing video models Models get chosen by reputation; films get ruined by it. Choose on evidence. ## Selection table (in script.md, at scripting time) Compare at least 3 video models the server exposes on: credits/second, reference & start/end-frame support, audio support, max clip duration, resolution. Note which judging criterion each strength serves. State a provisional pick per shot type. Identity-holding across shots usually matters more than anything else. ## The screen test (before shooting) 1. Take ONE approved board — the shot where identity is hardest to hold. 2. Generate a short clip of it with the top 2–3 candidates: same board, same prompt. 3. Extract 4 frames per clip; view them against the identity anchors. Which model keeps geometry/label/stack intact IN MOTION? 4. Record the comparison and verdict in `## Screen test` in script.md. 5. Commit. No mid-film model switch without a new screen test.
From maniculehq/creative-duel, MIT.
More from maniculehq/creative-duel
- 0.521score
concepting
maniculehq
Visual concept development on Nano Banana — moodboards, style frames, brand boards — and the script that comes out of it.
SkillBFreshness: BClaude Code +9 - 0.521score
magic-hour-api
maniculehq
Mechanical facts for using the magic-hour MCP server — uploads, polling, downloads, credit tracking.
SkillBFreshness: BClaude Code +9 - 0.518score
voice-audio
maniculehq
Casting voices without ears, and mixing the soundtrack like a professional.
SkillBFreshness: BClaude Code +9 - 0.518score
boards-identity
maniculehq
Locking your hero object's identity in validated stills before any video exists.
SkillBFreshness: BClaude Code +9 - 0.513score
dailies-qc
maniculehq
Reviewing every generated clip with your eyes — frame extraction, contact sheets, verdict rubric.
SkillBFreshness: BClaude Code +9 - 0.513score
cinematography-prompts
maniculehq
Prompt grammar that video models respond to — shots, moves, lighting continuity.
SkillBFreshness: BClaude Code +9 - 0.513score
edit-ffmpeg
maniculehq
Assembling the final 15.0s cut — concat, transitions, overlaid text, mux, and hitting spec.
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Aug 9, 2026 (2 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/maniculehq/creative-duel
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from maniculehq
- 0.521score
concepting
maniculehq
Visual concept development on Nano Banana — moodboards, style frames, brand boards — and the script that comes out of it.
SkillBFreshness: BClaude Code +9 - 0.521score
magic-hour-api
maniculehq
Mechanical facts for using the magic-hour MCP server — uploads, polling, downloads, credit tracking.
SkillBFreshness: BClaude Code +9 - 0.518score
voice-audio
maniculehq
Casting voices without ears, and mixing the soundtrack like a professional.
SkillBFreshness: BClaude Code +9 - 0.518score
boards-identity
maniculehq
Locking your hero object's identity in validated stills before any video exists.
SkillBFreshness: BClaude Code +9 - 0.513score
dailies-qc
maniculehq
Reviewing every generated clip with your eyes — frame extraction, contact sheets, verdict rubric.
SkillBFreshness: BClaude Code +9 - 0.513score
cinematography-prompts
maniculehq
Prompt grammar that video models respond to — shots, moves, lighting continuity.
SkillBFreshness: BClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get screen-test running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
screen-test skill. Agent Mart. https://agentmart.dev/skills/maniculehq-screen-test (last updated 2026-10-24)