- Home
- Skills
- Technical Documentation
- PRD to Execution Issues
PRD to Execution Issues
Turn a product spec into assignable engineering issues with clear completion checks and delivery order.
Free
Works with the AI Agents you already use
PRD to Execution Issues
Example session with this skill installed
Plan issues for article bookmarks: signed-in users can save, unsave, and view saved articles. Auth already exists. No folders or sharing. Return a Markdown issue plan.
- Read your context and instructions
- Compiled the prd to execution
Split article bookmarks into two testable delivery issues. Existing authentication is reused; folders and sharing are excluded.
Article Bookmarks Issue Plan
| Issue | Scope and boundaries | Acceptance criteria | Verification |
|---|---|---|---|
| B-1: Save and unsave articles | Add per-user bookmark storage, authenticated save/remove actions, and an article-page toggle. Excludes the saved list. | Saving survives refresh. Repeat saves create no duplicates. Unsave removes the bookmark. Signed-out requests are rejected. Users cannot change another user’s bookmarks. | Test save/remove persistence, duplicate saves, and access checks. Manually toggle and refresh an article. |
| B-2: Display saved articles | Add a saved-articles view using B-1’s storage and current-user lookup. Excludes folders and sharing. | Only the current user’s saved articles appear. Each opens its article. An empty list has a clear message. Removed bookmarks disappear on reload. | Test isolation with two accounts, empty state, navigation, and the list after removal. |
Sequence and ownership
- B-1: Medium complexity. Areas: data, API, article UI. No dependencies.
- B-2: Medium complexity. Areas: API, saved-list UI. Depends on B-1. Keep serial because storage and lookup behavior are shared.
Open question: What should happen when a saved article is deleted? Resolve before finalizing list behavior.
Next steps
- Confirm deleted-article behavior.
- Map work areas to repository modules.
- Assign B-1, then verify it before starting B-2.
Connects securely to your tools. The creator never sees your data.
About this skill
A product spec can explain what to build without giving your team clear tasks to start. Turn it into scoped engineering issues with testable outcomes, dependencies, and a practical delivery order.
What it does
- Creates assignable issues: Gives each task enough context for an engineer or coding agent to begin.
- Defines completion: Adds acceptance criteria and suggested tests or checks.
- Plans delivery order: Shows which tasks depend on others and which can run in parallel.
- Flags uncertainty: Highlights missing decisions, implementation risks, and scope boundaries.
How it works
- Share your product requirements document, feature plan, or implementation brief.
- Include known constraints, exclusions, and relevant implementation details.
- Receive a structured issue plan with dependencies, complexity estimates, and verification steps.
- Resolve open questions before assigning affected work.
Works with
Produces JSON issue plans suitable for Codex, GitHub Issues, Linear, and other engineering task systems. You can also request another output format. Direct issue creation is not included.
Why use this skill
Spend less time rewriting requirements into tickets and explaining what counts as finished. Tasks focus on testable behavior rather than disconnected database, API, and interface chores. Explicit dependencies help avoid assigning work before its prerequisites are ready.
Use cases
- Prepare a feature brief for a coding agent.
- Divide a product spec into assignable engineering work.
- Identify work that can proceed in parallel.
- Expose unresolved decisions before implementation starts.
Known limitations
This skill plans work, not implementation. Missing requirements can leave blocking questions that need your team’s answers.
Produces issue plans, not code or published tickets. Missing requirements may leave blocking questions that must be resolved before affected work begins.
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 1 month ago
- Free with an Agensi account
Permissions
May read a local PRD or repository context. Creates a structured issue plan in the response; does not write files or post to GitHub Issues or Linear by itself.