- Home
- Skills
- Business & Operations
- Feature Proposal Writer
Works with the AI Agents you already use
Feature Proposal Writer
Convert rough task descriptions into formal proposal.md and spec.md artifacts with GIVEN/WHEN/THEN criteria.
Free
Feature Proposal Writer
Example session with this skill installed
Scope a new feature for a CSV export button on the user management table. Use the specs/csv-export/ directory.
- Read your context and instructions
- Compiled the feature proposal writer
I have analyzed the requirements for the CSV export feature. Based on the user management context, I have generated the proposal and specification files.
Feature Artifacts
proposal.md
# Proposal: CSV Export for User Management
## Problem Statement
Administrators currently cannot extract user data for external reporting or auditing without direct database access.
## Proposed Solution
Add a "Export to CSV" action to the User Management table that generates a file containing the currently filtered view of users.
## Scope
- Frontend: Export button above the data table.
- Backend: Streamed CSV generation to prevent memory exhaustion.
- Data: ID, Name, Email, Role, and Created Date fields.
## Impact
Minimal impact on existing database schema. Potential load increase on the API during large exports.
## Verdict
PASS
spec.md
# Specification: CSV Export
## Requirements
- The system SHALL provide a button labeled "Export CSV" on the User Management page.
- The system MUST apply current table filters to the exported dataset.
- The export process SHALL NOT block the main UI thread.
## Acceptance Criteria
### Scenario 1: Basic Export
GIVEN an administrator is on the User Management page
WHEN they click the "Export CSV" button
THEN a file named `users_export_[timestamp].csv` is downloaded.
### Scenario 2: Filtered Export
GIVEN the administrator has filtered the table to "Role: Admin"
WHEN they click the "Export CSV" button
THEN the resulting file MUST only contain users with the Admin role.
### Scenario 3: Empty State
GIVEN the table has no data due to active filters
WHEN the administrator clicks the "Export CSV" button
THEN the system SHALL display a warning toast "No data available to export".
## Technical Constraints
- Must use stream-based CSV library for Node.js.
- Max export limit: 10,000 rows per request.
## Known Risks
- Large exports may timeout if the connection is slow. <TBD: Performance benchmarks>
Next steps
- Review the technical constraints in
spec.mdfor alignment with current backend limits. - Approve the
proposal.mdto move into the implementation phase.
Connects securely to your tools. The creator never sees your data.
About this skill
Vague task descriptions lead to scope creep and architectural debt. This skill enforces discipline by converting rough feature ideas into structured technical artifacts. It produces a clear rationale in proposal.md and a binding contract in spec.md before a single line of code is written.
What it does
- Requirement Analysis extracts core intent from task descriptions or existing project manifests.
- Proposal Generation builds a
proposal.mdfocusing on feature scope, technical impact, and necessity. - Specification Drafting creates a
spec.mdusing formal SHALL/MUST requirements. - Acceptance Criteria defines at least three GIVEN/WHEN/THEN scenarios for QA and testing.
- Architectural Guardrails flags proposals that violate project rules with a VETO response.
How it works
- Analyze Context by reading the task description and project README for domain constraints.
- Verify Input to ensure the description provides enough detail to avoid generic output.
- Generate Artifacts by populating standard templates for the proposal and technical spec.
- Identify Gaps by marking missing metrics or stakeholders with
<TBD>rather than inventing data.
Frameworks & tools
This tool is framework-agnostic and works with any codebase using Markdown-based documentation. It follows formal requirement engineering standards like RFC-style SHALL/MUST language.
Why this beats prompting it yourself
Generic prompts often hallucinate user metrics or ignore existing project architecture. This skill forces strict adherence to templates, mandates GIVEN/WHEN/THEN criteria, and provides a binary PASS/VETO verdict based on technical feasibility.
Use cases
- Drafting a technical design document for a new API endpoint.
- Defining acceptance criteria for a frontend component before development.
- Scoping a feature request from a non-technical stakeholder into a developer-ready spec.
Known limitations
This skill does not generate implementation code or file-level design diagrams. It is limited to producing exactly two Markdown files per feature.
Does not write code or design diagrams. Requires a task description to function. Outputs exactly two files: proposal.md and spec.md.
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
1 use
Times used to date
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 19 days ago
- Free with an Agensi account
Requires
Permissions
Reads the task description and project manifest; writes proposal.md and spec.md