Add hook evaluator — first judge in judges/ collection - #1
Open
rezakhadjavi wants to merge 3 commits into
Open
Conversation
Converts the hook-eval MCP into a self-contained skill. The evaluator classifies hooks as PASS/FAIL using psychological trigger analysis calibrated against real thumbstop performance data. The judges/ folder is designed to grow with additional evaluation skills over time. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Replaces the buried **Note:** CTAs with a consistent "With Real Data" section at the end of each action skill. Each section is contextual to what that skill just helped with, names the specific limitation (no real performance data), and positions Motion + the upcoming MCP as the natural next step. Applied to: hook-writing, ad-concept-generator, ugc-scriptwriter, judges/hook-evaluator. Skipped reference-only skills (hook-tactics, hook-voice-patterns) since they're libraries, not workflows. Also adds motion-product-context.md as a shared source of truth for Motion's platform capabilities and the upcoming MCP server. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Strip "Coming soon" MCP teaser from all skill files and the shared product context. We'll add these back when the Motion MCP is live. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
judges/hook-evaluator/— a skill that evaluates ad hooks for scroll-stopping power, returning PASS/FAIL with psychological trigger analysis and improvement suggestionsjudges/folder as a new category that can grow with additional evaluation skills over time (concept judge, script judge, etc.)What changed
judges/hook-evaluator/SKILL.md— the evaluator skillmarketplace.json— registers the new skillREADME.md— adds the evaluator to the skills table, example prompts, and pipeline flowHow it fits the pipeline
Hooks Writing → Hook Evaluator (quality gate) → Ad Concept → UGC Script
The evaluator sits between generation and concept development as a quality filter. Users generate hooks, evaluate them, rewrite failures using the suggestions, and only develop passing hooks into concepts.
Test plan
🤖 Generated with Claude Code