cross-platform-dev-scripts
Skill2 starsChanged 17 days agoApache-2.0View on GitHub
Generates and validates turnkey, cross-platform dev setup and test scripts (scripts/win and scripts/linux) with dual-engine uv/venv support. Use when setting up a repository for local onboarding, adding Windows and Linux dev scripts, or verifying script parity.
- Deletes or force-pushes
In this repository19 other files it gives its agents.
- .agents/AGENTS.mdAGENTS.md
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- lessons-learntSkill
- multi-agent-docsSkill
- preflight-test-engineerSkill
- public-repo-release-reviewSkill
- readme-designerSkill
- saas-app-builderSkill
- third-party-skill-verifierSkill
- skill-creatorSkill
- cross-platform-dev-scriptsSkill
- lessons-learntSkill
- multi-agent-docsSkill
- preflight-test-engineerSkill
- public-repo-release-reviewSkill
- readme-designerSkill
- saas-app-builderSkill
- third-party-skill-verifierSkill
The file
We split this skill into the rules it gives an agent
8 rules. About 0 are common, 2 are distinctive, and 6 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 30
Never mandates external downloads or pipes unverified scripts into a shell.
- Very specific: line 34
Never requires sudo or Administrator escalation.
- Very specific: line 35
All paths and cleanup routines (rm -rf -- "$VENV_DIR") are strictly quoted.
- Very specific: line 40
Test filtering: supports -k <expr> (Bash) and -K <expr> (PowerShell) for fast local iteration.
- Very specific: line 64
If the project requires multiple requirements files (e.g.
- Very specific: line 66
If tests require specific local environment flags (e.g.
- Very specific: line 68
If the project has linting or type-checking steps (e.g.
- Distinctive: line 73
Run the verification tool to audit the created scripts for safety invariants and cross-platform parity:
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Cross-Platform Dev Scripts & Local Onboarding
Purpose
Provides a standardized, atomic workflow to equip any repository with hardened, turnkey developer setup and test scripts:
scripts/linux/dev-setup.sh(Linux / macOS / WSL)scripts/linux/dev-test.shscripts/win/dev-setup.ps1(Windows PowerShell 5.1 / 7+)scripts/win/dev-test.ps1
Empowers any contributor or autonomous agent to clone a repository and immediately run setup and test commands with zero configuration friction.
Architectural Principles
- Dual-Engine Acceleration (
uv+venv):- Uses
uvwhen installed on the user's PATH for 10x–100x faster package resolution. - Automatically and gracefully falls back to the Python standard library's
venv+pipwhenuvis absent. - Never mandates external downloads or pipes unverified scripts into a shell.
- Uses
- Strict Hermetic Isolation (Least Privilege):
- Everything is confined strictly to
.venv/at the repository root. - Never requires
sudoor Administrator escalation. - All paths and cleanup routines (
rm -rf -- "$VENV_DIR") are strictly quoted.
- Everything is confined strictly to
- Parity & Developer Experience (DX):
- Identical test targets and step sequences between Windows PowerShell and Linux Bash.
- Auto-bootstrap:
dev-testrunsdev-setupautomatically if the virtual environment does not exist yet. - Test filtering: supports
-k <expr>(Bash) and-K <expr>(PowerShell) for fast local iteration.
Procedural Workflow
Phase 1: Heavy-Lifting Scaffolding
Inspect the target project root and execute the built-in generator script:
python <skill_path>/scripts/scaffold_dev_scripts.py --target . [--requirements requirements-dev.txt] [--test-cmd "-m pytest"]The script automatically:
- Detects project manifests (
requirements-dev.txt,requirements.txt,pyproject.toml, etc.). - Creates
scripts/linux/andscripts/win/directories idempotently. - Generates the 4 hardened scripts with proper permissions (
0o755for.shscripts). - Outputs a JSON summary of generated files.
Phase 2: Semantic Tailoring (Agent Review)
Inspect the generated scripts and tailor them to any project-specific constraints:
- Requirements & Build Tools:
- If the project requires multiple requirements files (e.g.
requirements.txt+requirements-dev.txt), update the setup scripts to install both.
- If the project requires multiple requirements files (e.g.
- Environment Variables & Secrets:
- If tests require specific local environment flags (e.g.
TESTING=1orPYTHONPATH=.), configure them indev-test.shanddev-test.ps1.
- If tests require specific local environment flags (e.g.
- Multi-Step Test Pipelines:
- If the project has linting or type-checking steps (e.g.
ruff check,mypy), add correspondingrun_step(Bash) andInvoke-Step(PowerShell) invocations in both test scripts in identical order.
- If the project has linting or type-checking steps (e.g.
Phase 3: Parity & Safety Verification
Run the verification tool to audit the created scripts for safety invariants and cross-platform parity:
python <skill_path>/scripts/validate_dev_scripts.py --target .The validator verifies:
- Script presence and shebang formatting.
- Strict error handling (
set -Eeuo pipefailin Bash,$ErrorActionPreference = "Stop"in PowerShell). - Safe quoting and isolation (no
sudoinvocations, no piping remote scripts to a shell, no writes outside.venv/). - Cross-platform step parity and filter support.
- Static syntax checking (
bash -nand PowerShell AST parsing when tooling is present).
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

