- Home
- Skills
- Technical Documentation
- codex-spec-interviewer
More screenshots
Works with the AI Agents you already use
Codex Spec Interviewer
Turn fuzzy requests into verified implementation specs and ADRs with source-backed architectural gating.
Free
codex-spec-interviewer
Example session with this skill installed
We need to update our auth to support JWT rotation. Check our security policy and existing middleware patterns, then prepare the spec and any necessary ADRs for the implementation.
- Read your context and instructions
- Compiled the codex-spec-interviewer
Artifacts saved: docs/specs/auth-refactor-spec.md, docs/adrs/0004-jwt-rotation.md
Summary: Verified JWT rotation strategy vs repo security policy.
Validation: Run npm test and scripts/verify-auth.sh.
Execution Prompt: "Use auth-refactor-spec.md to update the middleware..."
Connects securely to your tools. The creator never sees your data.
About this skill
Stop Coding in the Dark
The Codex Spec Interviewer is a rigorous requirements-engineering skill for AI agents. It bridges the gap between a vague user request ("Add a billing dashboard") and a production-grade implementation plan that a coding agent can actually execute without hallucination or scope creep.
What it does
This skill engages in a "source-backed interview," scanning your repository's local AGENTS.md, existing ADRs, and codebase to challenge assumptions before a single line of code is written. It automatically gates architectural changes, ensuring significant decisions are captured as ADRs (Architecture Decision Records) rather than buried in chat history.
Supported Workflows
- Compact: Fast-track for low-risk, small feature edits.
- Standard: The default for bugfixes, refactors, and features.
- Deep: For complex, repo-wide migrations and architectural shifts.
Why use this skill?
Prompting an AI to "just write the code" often leads to technical debt and missed edge cases. This skill ensures your agent outputs a durable -spec.md file including testable acceptance criteria, validation commands, and rollout risks. It provides a dedicated 'Codex execution prompt'—a high-context handoff instructions specifically designed for coding agents to follow predictably.
- Requires existing repo documentation (like AGENTS.md) for best results.
- Not intended for simple, one-line code edits.
- Relies on tool-calling for file persistence.
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
9 uses
Times used to date
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 5 months ago
- Free with an Agensi account