- Home
- Skills
- Prompt & Skill Engineering
- nexus kit
Works with the AI Agents you already use
nexus kit
Enforce Spec-Driven Development with a multi-phase orchestrator for planning, coding, and verifying requirements.
Free
nexus kit
Example session with this skill installed
I need to add a secure password reset flow to our TypeScript API. Use SDD to plan and implement this.
- Read your context and instructions
- Compiled the nexus kit
I have initialized the .specify/ directory and generated a draft specification for the password reset flow. I have focused on security edge cases and the token expiration logic.
Task Specification: 001-password-reset-flow.md
| Section | Details |
|---|---|
| Status | Draft (v1.0.0) |
| In-Scope | Token generation (SHA-256), Email dispatch interface, Password hashing (Argon2), Rate limiting |
| Out-of-Scope | Front-end UI components, SMS MFA, Social login recovery |
| Domain Models | ResetToken { userId: UUID, hash: string, expiresAt: Date, used: boolean } |
| AC-1 | Generate unique 32-character tokens with 1-hour expiration |
| AC-2 | Invalidate all existing tokens for a user when a new one is requested |
| AC-3 | Return generic 200 OK for non-existent emails to prevent account enumeration |
| Edge Cases | Expired token usage, Rapid-fire request throttling, Replay attacks |
Next steps
- Review the full spec in
.specify/tasks/001-password-reset-flow.md. - Approve the spec by changing status to
Approvedor runningnpx create-nexus-kit@latest --approve. - Signal me to begin the
sdd-coderphase once approved.
Connects securely to your tools. The creator never sees your data.
About this skill
Vibe coding leads to architectural drift, missing edge cases, and brittle implementations. This skill enforces disciplined Spec-Driven Development (SDD) by requiring a verified technical specification before any source code is touched.
What it does
- Enforce SDD governance by mandating a "No Spec, No Code" policy for all features and refactors.
- Automate spec generation using standardized templates that include domain models, logic flows, and edge cases.
- Control scope creep by explicitly defining in-scope and out-of-scope boundaries before implementation begins.
- Verify implementations through a traceability matrix that maps every acceptance criterion to specific code and tests.
- Manage requirement drift via a formal versioning system and change log for approved specifications.
How it works
- Bootstrap: Initializes the
.specify/directory with your project constitution and standardized spec templates. - Plan: Generates a detailed draft specification and stops for human approval before proceeding.
- Implement: Writes type-safe, modular code and automated tests based strictly on the approved spec.
- Verify: Maps test results to acceptance criteria and marks the task as completed only when 100% of requirements pass.
Frameworks & tools
The skill is framework-agnostic but optimized for type-safe environments (TypeScript/Rust/Go). It integrates with the nexus-kit CLI for SHA-256 spec verification and repository integrity checks.
Why this beats prompting it yourself
Generic prompts often ignore edge cases or hallucinate extra features that violate your architecture. This skill uses a multi-phase orchestrator (Planner, Coder, Verifier) to ensure every line of code is backed by a signed requirement and a passing test.
Use cases
- Designing new API modules or UI components with strict contract requirements.
- Planning complex refactors to ensure zero regression of existing business logic.
- Building high-stakes features where edge-case handling and error schemas are mandatory.
- Auditing existing codebases against a set of technical specifications.
Known limitations
Requires a manual approval step between planning and implementation. Best suited for greenfield features or isolated modules rather than bulk legacy code migrations.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Prompt & Skill Engineering.
Requires human approval at the transition from planning to implementation. Optimized for projects that support structured testing and type safety.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 7 days ago
- Free with an Agensi account
- Software
- create-nexus-kit
Node.js
Permissions
No special permissions declared or detected