unknowns-spec-interview
Skill2 starsChanged 4 months agoMITView on GitHub
Interview the user one question at a time about ambiguous requirements, ordered by blast radius. Use when the user says "interview me", "do not assume", wants one question at a time, has ambiguous specs, or needs Codex to identify which answers would change architecture, scope, data model, UX, or rollout.
In this repository11 other files it gives its agents.
- know-your-unknownsSkill
- unknowns-blindspot-passSkill
- unknowns-buy-in-docSkill
- unknowns-design-directionsSkill
- unknowns-domain-bootstrapSkill
- unknowns-implementation-notesSkill
- unknowns-intervention-brainstormSkill
- unknowns-merge-quizSkill
- unknowns-mock-before-wireSkill
- unknowns-reference-mapSkill
- unknowns-tweakable-planSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 10
Ask the few questions that matter most.
- Distinctive: line 16
Ask one concise question at a time.
- Distinctive: line 17
After answers, produce a decision table and implementation prompt.
- Distinctive: line 25
Stop condition: no high-blast-radius ambiguity remains.
- Distinctive: line 29
Do not ask a broad questionnaire.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Unknowns Spec Interview
Workflow
Ask the few questions that matter most.
Process:
- Identify the unresolved decisions.
- Rank them by blast radius: architecture, data model, API contract, user-facing behavior, rollout, then copy/details.
- Ask one concise question at a time.
- After answers, produce a decision table and implementation prompt.
Return after the interview:
- Decisions made.
- Decisions still open.
- Assumptions Codex is allowed to make.
- Ready-to-paste implementation prompt.
- Stop condition: no high-blast-radius ambiguity remains.
Rules
Do not ask a broad questionnaire. Do not ask low-impact polish questions while architecture-shaping questions remain. If the user answers indirectly, restate the decision before proceeding.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.