From project-toolkit
Enforces taste invariants via custom lints: file size limits, naming conventions, function complexity, skill prompt size, with agent-readable remediation instructions. Use when writing or reviewing code.
npx claudepluginhub rjmurillo/ai-agents --plugin project-toolkitThis skill uses the workspace's default tool permissions.
Custom lints where error messages become agent-readable remediation instructions.
Generates design tokens/docs from CSS/Tailwind/styled-components codebases, audits visual consistency across 10 dimensions, detects AI slop in UI.
Records polished WebM UI demo videos of web apps using Playwright with cursor overlay, natural pacing, and three-phase scripting. Activates for demo, walkthrough, screen recording, or tutorial requests.
Delivers idiomatic Kotlin patterns for null safety, immutability, sealed classes, coroutines, Flows, extensions, DSL builders, and Gradle DSL. Use when writing, reviewing, refactoring, or designing Kotlin code.
Custom lints where error messages become agent-readable remediation instructions.
Inspired by OpenAI Harness Engineering:
"We statically enforce structured logging, naming conventions for schemas and types, file size limits, and platform-specific reliability requirements with custom lints. Because the lints are custom, we write the error messages to inject remediation instructions into agent context."
| Trigger Phrase | Operation |
|---|---|
run taste lints | taste_lints.py on staged or specified files |
check file size | taste_lints.py with file-size rule only |
check naming conventions | taste_lints.py with naming rule only |
lint taste invariants | taste_lints.py full scan |
taste lint report | taste_lints.py with JSON output |
Use this skill when:
Files exceeding line thresholds indicate poor cohesion.
| Threshold | Lines | Severity |
|---|---|---|
| Warning | 301-500 | warning |
| Error | 501+ | error |
| Pattern | Rule | Applies To |
|---|---|---|
snake_case | Python files, functions, variables | *.py |
kebab-case | Skill directories, YAML files | .claude/skills/, *.yml |
PascalCase | PowerShell functions, classes | *.ps1, *.psm1 |
UPPER_CASE | Constants, environment variables | All languages |
invoke_ prefix | Hook scripts | .claude/hooks/ |
Functions exceeding cyclomatic complexity 10 need decomposition.
Skills exceeding 500 lines need progressive disclosure refactoring.
python3 .claude/skills/taste-lints/scripts/taste_lints.py with target files# Scan staged files
python3 .claude/skills/taste-lints/scripts/taste_lints.py --git-staged
# Scan specific files
python3 .claude/skills/taste-lints/scripts/taste_lints.py path/to/file.py
# Scan a directory
python3 .claude/skills/taste-lints/scripts/taste_lints.py --directory src/
# JSON output
python3 .claude/skills/taste-lints/scripts/taste_lints.py --format json --git-staged
# Run specific rules only
python3 .claude/skills/taste-lints/scripts/taste_lints.py --rules file-size,naming
| Code | Meaning |
|---|---|
| 0 | No violations found |
| 1 | Script error (bad arguments, file not found) |
| 10 | Violations detected |
Add a comment to suppress a specific rule on a file:
# taste-lint: ignore file-size
Valid rules: file-size, naming, complexity, skill-size