product-manager

Converts product ideas into BRDs with user stories, acceptance criteria, and compliance scoping.

1|Updated Jun 24, 2026
One-click install
npx skills add https://github.com/sundarshahi/drydock --skill product-manager-sundarshahi
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: product-manager
Source: https://github.com/sundarshahi/drydock/tree/main/skills/product-manager
Command: npx skills add https://github.com/sundarshahi/drydock --skill product-manager-sundarshahi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Turning a vague product idea into engineering-ready requirements is slow and error-prone. This Skill interviews the stakeholder, researches the domain, and produces a formal BRD with testable acceptance criteria, then verifies that the implemented code actually matches those requirements. ## Core Features & Use Cases - Adaptive CEO Interview: Asks 2-12 structured questions depending on autonomy level (Autopilot, Copilot, Checkpoint, Manual), with a mandatory compliance and data-classification discovery question at every level. - BRD Authoring & Management: Writes a canonical brd.md with user stories, acceptance criteria, business rules, and a resolved compliance scope, plus a living INDEX.md tracking feature status. - Autonomous Verification: Spawns a verification agent that checks each acceptance criterion against the codebase and reports PASS/FAIL/PARTIAL per criterion. - Use Case: Describe a new SaaS feature in plain English; the Skill interviews you, drafts the BRD at drydock/product-manager/BRD/brd.md, gets your approval, and later confirms the engineering work matches every criterion. ## Quick Start Ask the product-manager skill to turn your feature idea into a BRD with user stories and acceptance criteria.

Frequently Asked Questions about product-manager

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I turn a product idea into a BRD with AI?▼

Describe your feature idea in plain English and the Skill runs a structured interview, researches the domain, then writes a BRD with user stories, acceptance criteria, and business rules at drydock/product-manager/BRD/brd.md. You approve the draft before it is handed to engineering.

How to write testable acceptance criteria for user stories?▼

Acceptance criteria follow the Given/When/Then format and must be specific and measurable, such as returning a 200 response with valid JSON within 500ms. The Skill rejects vague phrasing like works well and co-validates criteria with you at higher autonomy levels.

Does the product manager skill handle compliance requirements?▼

Yes, it asks a mandatory compliance and data classification question at every autonomy level and resolves a deterministic framework scope such as HIPAA, PCI-DSS, GDPR, CCPA, SOC 2, or FedRAMP into the BRD. Unsure answers default to a conservative in-scope set flagged for compliance-officer confirmation.

Can it verify that code matches the requirements?▼

Yes, Phase 4 spawns a verification agent that reads each BRD acceptance criterion and compares it against the actual implementation, reporting PASS, FAIL, or PARTIAL per criterion with an overall compliance percentage. The BRD status is then updated to Verified or Done.

When should I not use a product requirements skill?▼

Do not use it for pure technical tasks such as bug fixes or refactoring that do not change business logic. It writes specifications and verifies implementation but does not write code itself; implementation is handed off to engineering skills.