See it in action
One of 10 test cases the skill fixed. Same model, same prompt, with and without the skill.
PromptWhen verifying that newly scaffolded tests pass in a Python environment, what specific test invocation is executed?
What was checked, without and with the skill| Checked | Without skill | With skill |
|---|
| Runs pytest -v to verify tests pass | Missed | Passed |
|---|
What's inside
The instructions your model reads when the skill loads.
SKILL.md147 linessingle file
Project Initialization Skill
Interactive workflow for initializing new software projects with complete development infrastructure.
Use When
- Starting a new Python, Rust, or TypeScript project
- Updating existing project tooling to current standards
- Need to set up git, GitHub workflows, pre-commit hooks, Makefile
- Want consistent project structure across team
- Converting unstructured project to best practices
- Adding missing configurations to established codebases
Workflow
1. Detect or Select Language
Load modules/language-detection.md
- Auto-detect from existing files (pyproject.toml, Cargo.toml, package.json)
- If ambiguous or empty directory, ask user to select
- Validate language is supported (python, rust, typescript)
2. Collect Project Metadata
Load modules/metadata-collection.md
Gather:
- Project name (default: directory name)
- Author name and email
- Project description
- Language-specific settings:
- Python: version (default 3.10)
- Rust: edition (default 2021)
- TypeScript: framework (React, Vue, etc.)
- License type (MIT, Apache, GPL, etc.)
3. Review Existing Files
Check for existing configurations:
Track record
Benchmark
19 / 22
test cases passed, vs 9 / 22 without
Adoption
- Users
- 0
- Times used
- 0
Safety and source details
SkillSafety
✓ 3 of 3 passed✓Code scan15 of 15 checks clean
0 critical · 0 warning · 0 info — scanner v5 — 9/18/2026
View 15 scanner checks
live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing
✓IntentSafe
LLM judge · does the skill do what it claims
✓ContentClean
LLM judge · policy & harmful-content review
Source
Importedathola / claude-night-market
plugins/attune/skills/project-init/SKILL.md
License MITImported 9/18/2026Version v2