⏳ This skill is pending AI review.
Scores will appear once the review pipeline completes.
code-review-and-quality
Conducts multi-axis code review. Use before merging any change. Use when reviewing code written by yourself, another agent, or a human. Use when you need to assess code quality across multiple dimensions before it enters the main branch. Use when asked to review a diff or a pull request, even when the diff is pasted inline.
Choose how to use this skill
You do not need every option. Choose the path your AI client supports. The stable page stays the same; versioned files are immutable.
1. Native installer
This listing has no registered native installer command. Use the complete package or source fallback below, depending on what your client supports.
Do not guess an installer command or replace an existing version without reviewing the diff.
2. Complete package recommended
Download the ZIP when available. It includes SKILL.md plus the references, security notes and version metadata.
No complete ProSkills package is published for this listing yet.3. Prompt-only
Copy the prompt above when the agent can read the stable page or when you want to adopt the workflow without installing a skill.
Need only the instruction file?
Download SKILL.md only if your client requires a single file. The complete ZIP is safer for a full installation because it preserves the references and release context.
No path installs or executes anything by itself. Your agent still needs access to the project files. Before updating, compare the installed version and review the diff.
// RATINGS
Not yet listed on ClawHub or SkillsMP
// README
WalidForge
Understand your project. Make the change. Show the proof.
WalidForge brings backend, frontend, mobile and DevOps workflows into one portable skill pack. Define the outcome, choose the simplest suitable design, deliver in small slices, and verify the failure paths before calling the work complete.
Maintained by bnalwalid. Repository: bnalwalid/walidforge-skills.
Engineering defaults: document the architecture before implementation, choose boundaries from project constraints, and show evidence for the delivered behavior. All stacks are welcome.
Learn the workflow
Read the complete user guide for installation choices, all 25 skills, practical examples, Doctor/Compass usage, change review and troubleshooting. Start with the quick start for your first task.
Start with your project
From this checkout, inspect the project you want to work on:
node scripts/forge.js doctor /path/to/your-project
node scripts/forge.js doctor /path/to/your-project --json
Project Doctor reports observed stack markers, declared check commands, candidate commands and metadata conflicts. It uses Node built-ins with no dependencies, network calls, project-script execution or target writes. A discovered command is not a test result; review it against your README and CI before running it.
It recognizes common markers for JavaScript/TypeScript, PHP, Python, Go, Rust, Ruby, JVM, .NET, Dart/Flutter, Swift, containers and Terraform at the selected root. Nested packages and unsupported stacks need manual discovery.
Then ask your agent for the outcome you need: understand code, fix a failure, build a feature, improve a boundary, prepare a release or explain a concept. No catalog memorization is required. See the Developer Handbook for daily tasks, supported markers, command provenance and limits.
Plan a change with evidence
Change Compass links a feature or bug request to candidate source files, tests and possible direct consumers, then produces decision questions and a verification plan using Project Doctor command provenance:
node scripts/forge.js compass /path/to/project --intent "fix duplicate invoice" --kind bug
node scripts/forge.js compass /path/to/project --intent "add account permissions" --json
It runs locally without a model account or network. Matching is lexical: include
identifiers used by your project. Candidates require inspection, and no suggested
command is executed. See Change Compass for the contract,
coverage limits and what the report can establish. Save its JSON output before
implementation, then use --review --baseline /path/to/snapshot.json to compare
Git changes with the saved candidate paths. Review does not claim tests passed.
How WalidForge earns trust
- Project evidence before assumptions: work from the actual stack, owner and canonical commands, including legacy projects and monorepos.
- Proportionate workflows: keep routine changes focused; deepen checks for data, permissions, concurrency, device lifecycle and release risks.
- An outcome-to-evidence handoff: report what changed, the consequential decision, executed checks, remaining limits and the next unfinished action.
- Evaluated decisions: structural and routing checks plus realistic behavioral scenarios; report which tiers actually ran.
These are design choices you can inspect and test, not a claim that no other project shares them or that every platform has been validated.
Four engineering perspectives
| Domain | Decisions covered | Entry skill |
|---|---|---|
| Backend | Atomic invariants, tenant boundaries, bounded queries, durable delivery and compatible migrations | API and interface design |
| Frontend | Component boundaries, state ownership, stale requests, accessible interaction and runtime verification | Frontend UI engineering |
| Mobile | Offline intent, process recreation, account isolation, native interaction and installed-client compatibility | Native mobile delivery |
| DevOps | Infrastructure plans, privilege separation, artifact promotion, release signals and data-aware recovery | CI/CD and automation |
Use the project's existing stack. Load the relevant workflow when needed. A simple change should not introduce an elaborate architecture, and a risky change should not hide behind a green build.
Install
The commands below target bnalwalid/walidforge-skills. Choose an installation
that matches your host and verify which resources it carries.
For hosts supported by the skills installer:
npx skills add bnalwalid/walidforge-skills
For a complete local checkout, including shared references and adapters:
git clone https://github.com/bnalwalid/walidforge-skills.git
cd walidforge-skills
Per-skill installs may omit the shared references/ directory. Use a whole-repo
integration when a workflow needs those checklists. See
getting started,
Codex setup, and
other host commands.
The workflow
Define → Plan → Build → Verify → Review → Ship.
Lifecycle wrappers expose /spec, /plan, /build, /test, /review, /ship,
/constraints, /code-simplify and /webperf where the host supports them.
Other hosts discover the full skill names directly. Do not assume installation
of individual skills also installs command wrappers or shared reference files.
The pack includes 25 skills and four focused review personas. Published skill IDs
stay stable; plugin namespaces and marketplace identity use walidforge-skills.
What counts as done
Requirements are explicit. Decisions name their constraints and tra
// HOW IT'S BUILT
KEY FILES