- Home
- Skills
- Code Quality & Review
- requirements analysis
requirements analysis
Perform rigorous static testing on requirements to catch ambiguities and missing criteria before implementation.
$10
Works with the AI Agents you already use
requirements analysis
Example session with this skill installed
Analyze this user story: As a user, I want the search results to load immediately and show relevant items. AC:
- Search works.
- Results are displayed.
- Read your context and instructions
- Compiled the requirements analysis
I have completed the requirements analysis for the Search Results user story.
Requirements Analysis Report
| ID | Source quote/location | Category | Issue or testable criterion | Severity | Impact | Proposed clarification/rewrite |
|---|---|---|---|---|---|---|
| AC-1 | "load immediately" | Untestable | Missing measurable timing and network context. | Critical | Cannot pass/fail. | "Results display within 200ms on 4G." |
| AC-2 | "relevant items" | Ambiguity | "Relevant" is subjective; no sorting/ranking logic defined. | Major | Wrong data shown. | Define ranking by "Newest" or "Keyword Match". |
| AC-3 | "Search works" | Undefined | Does not specify triggers (Enter key, icon click, or type-ahead). | Major | Implementation gap. | "Search triggers after 3 chars or on Enter." |
| GAP-1 | N/A | Missing | No zero-results state defined. | Minor | UX dead end. | Add AC for "No results found" message. |
Questions
- Business rule: What is the maximum number of results to display per page?
- UI/flow: Should the search filter apply as the user types or only after submission?
- Data/integration: Which specific database fields are included in the "relevance" calculation?
Assumptions (unconfirmed)
- Search is restricted to the primary product catalog only.
- User is already authenticated before accessing search.
Readiness: NEEDS CLARIFICATION
The requirement is blocked by untestable performance targets ("immediately") and undefined relevance logic.
Next steps
- Review the proposed rewrites with the Product Owner to define "relevance."
- Define the technical latency threshold (e.g., 200ms or 500ms).
- Document the "No Results" behavior.
Connects securely to your tools. The creator never sees your data.
About this skill
Vague user stories and incomplete acceptance criteria lead to scope creep and expensive late-stage bug fixes. This skill performs rigorous static testing on your requirements to catch ambiguities before a single line of code is written.
What it does
- Ambiguity detection identifies unverifiable terms like "fast" or "appropriate" that prevent clear pass/fail outcomes.
- Gap analysis uncovers missing boundary conditions, error states, and undefined business rules in your specifications.
- Traceability mapping builds a direct link from source requirements to testable outcomes across functional and non-functional domains.
- Criteria refinement generates proposed rewrites for acceptance criteria to ensure they are measurable and actor-oriented.
- Readiness assessment provides a structured GO/NO-GO disposition based on unresolved blockers and implementation risks.
How it works
- Extraction of actors, goals, triggers, and data flows from your raw text or project files.
- Static review against ISO/IEC 25010 quality characteristics to find conflicts or untestable wording.
- Question generation where the agent asks prioritized questions categorized by business rules, UI, and integration needs.
- Final reporting using a standardized table format that includes impact scores and proposed clarifications.
Frameworks & tools
The skill follows ISTQB static testing principles and ISO/IEC 25010 quality standards. It is optimized for use within Claude Code, Cursor, or ChatGPT when analyzing Markdown-based requirements, Jira exports, or PRD documents.
Why this beats prompting it yourself
Generic prompts often miss non-functional requirements or silently hallucinate product behavior to fill gaps. This skill enforces a strict evidence-readiness gate, ensuring every suggestion is traceable to your source text rather than an LLM guess.
Use cases
- Reviewing a new PBI or User Story before a sprint planning session.
- Analyzing UI mockups to find missing validation messages and error states.
- Auditing a change request for impacts on existing system dependencies.
- Performing a formal peer review on technical specifications or API contracts.
Known limitations
Does not replace stakeholder decisions or provide legal compliance guarantees. Requires a writable folder for saving reports if used in a local CLI environment.
Requires user confirmation for output folders. Cannot resolve business conflicts without stakeholder input. Does not guarantee compliance with external standards.
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 2 days ago