Install
$ agentstack add skill-akovalion-paranoid-qa-interview ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Interview: requirements gathering through questions
Help the user turn an informal task description into complete requirements ready to hand off to the test-cases or testing skills.
When to apply
- The task is described verbally or in general terms, without a ticket
- No AC, no mockups, no clear scope
- The user is unsure what exactly to test/automate
If the task already comes from an issue tracker (Jira etc.) with a full description — this skill is not needed: gather context straight from the ticket (via the tracker's MCP: description, comments, attachments, linked issues) and hand off to test-cases.
Process
Round 1: scope and context. Ask 3-4 questions via AskUserQuestion:
- What exactly is being checked (feature, page, element)
- Where it lives (URL, screen, navigation path)
- Task type (new functionality / regression / bug)
- Platforms (desktop / mobile / both)
Round 2: behavior and AC. Based on the round 1 answers:
- Happy path — what behavior counts as correct
- Edge cases — which negative scenarios matter
- UI states (loading, error, empty, disabled) — if applicable
- Integrations — any APIs / external services
Round 3: clarifications on remaining gaps. Only if critical gaps remain after rounds 1-2. Group related questions.
Maximum 3 rounds. If the requirements are still incomplete — log the gaps as "Questions for the analyst/PM" and work with what you have.
Output format
Once the interview is done, assemble a structured summary:
# Requirements:
## Context
- Type:
- Platforms:
- URL / location:
## Scope
## Acceptance Criteria
-
-
## Scenarios
**Happy path:**
**Edge cases:**
-
-
## Integrations
## Open questions for the analyst/PM
-
Save the summary to an md file (e.g. requirements_.md in the current working directory) and offer to hand it off to test-cases or testing.
What to prefer
- 4 precise questions beat 10 boilerplate ones
- Tie questions to the previous round's answers instead of running down a checklist
- If the user answers "don't know" — log it as a question for the analyst, do not push
- Use concrete answer options in questions (AskUserQuestion options), not open-ended wording
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: akovalion
- Source: akovalion/paranoid-qa
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.