- Home
- Skills
- Agents & Orchestration
- Implementation Blueprint
Implementation Blueprint
Converts feature requests into scoped implementation plans with dependency-ordered tasks and AI-builder prompts.
$2.99
Works with the AI Agents you already use
Implementation Blueprint
Example session with this skill installed
Plan the implementation for a new 'Password Reset' flow. I have a Users table and an existing SES email adapter. Need to handle token expiration and rate limiting.
- Read your context and instructions
- Compiled the implementation blueprint
I have analyzed the request for a Password Reset flow. Below is the implementation blueprint centered on the existing Users table and SES adapter, prioritizing token security and rate limiting.
Implementation Blueprint: Password Reset Flow
1. Technical Strategy
We will use a short-lived PasswordResetToken table to decouple reset logic from the primary Users entity. This avoids bloating the user schema and allows for easy cleanup of expired tokens.
| Task | Scope | Dependencies | Acceptance Criteria |
|---|---|---|---|
| 01: Schema | Create password_reset_tokens table. | None | Table exists with user_id, hashed_token, and expires_at. |
| 02: Token Gen | Logic to generate 32-char secure token. | 01 | Token is hashed before storage; plain text never logged. |
| 03: Rate Limit | Middleware for /api/reset-request. | None | Max 3 requests per IP per hour; returns 429 on breach. |
| 04: Mailer | Integrate SES with reset template. | 02 | Email contains unique URL; SES 4xx/5xx errors handled. |
| 05: Reset UI | Form to submit new password via token. | 01, 02 | Validates token exists and now() < expires_at. |
2. Builder Prompt: Task 02 (Token Generation)
Context: Implement the business logic for password reset token generation.
Files: src/lib/auth/tokens.ts, src/db/schema.sql
Constraints:
- Use crypto.randomBytes for entropy.
- Set expiration to 1 hour from generation.
- Ensure any existing tokens for the same user_id are invalidated first.
Verification:
- Assert that the plain text token is returned to the caller but the SHA-256 hash is stored.
- Provide a unit test for expiration logic.
Next steps
- Provide the current schema for the
Userstable to finalize the foreign key relationship. - Confirm if the SES adapter uses a specific template ID or raw HTML.
- Review the rate limiting thresholds for production environment alignment.
Connects securely to your tools. The creator never sees your data.
About this skill
Feature requests often fail because they lack technical depth or concrete execution steps. This skill forces clarity by converting vague requirements into a structured implementation blueprint. It stops the cycle of half-baked features by identifying dependencies, data contract changes, and failure handling before a single line of code is written.
What it does
- Scope definition clarifies outcomes, non-goals, and constraints to prevent scope creep.
- Dependency mapping orders build tasks based on prerequisites and data contracts.
- Trade-off analysis compares technical alternatives based on maintenance and operating costs.
- Task breakdown generates vertical slices with objective acceptance criteria and verification steps.
- Builder prompts produces copyable AI-ready prompts for specific code generation tasks.
How it works
- Discovery where the agent asks up to three targeted questions to resolve ambiguity.
- Context inspection to analyze existing modules, interfaces, and conventions before proposing integrations.
- Drafting the blueprint with scoped options, architectural recommendations, and deferred work.
- Verification of every task slice to ensure acceptance criteria are met within the allowed scope.
Frameworks & tools
This skill is framework-agnostic. It works by analyzing your existing repository structure and conventions. It focuses on logic, data contracts, and API semantics rather than specific library versions.
Why this beats prompting it yourself
Manual prompting often results in "hallucinated" project paths or tasks that ignore existing dependencies. This skill uses a formal planning workflow that mandates evidence-based inspection and explicit failure handling policies, ensuring your plan is actually buildable.
Use cases
- Map out a new API endpoint including validation, storage, and error handling.
- Plan a complex frontend feature with mocked state before backend completion.
- Design resilient external service integrations with explicit retry and idempotency logic.
- Break down a monolithic PR into small, verifiable vertical slices.
Known limitations
Does not perform code edits, installations, or deployments. Requires manual inspection of relevant files to avoid generic output. No time or cost estimates are provided.
Cannot execute code, deploy, or access secrets. Requires user-provided file context to avoid generic plans. Does not provide time or cost estimates.
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 any payment.
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
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 10 days ago