prd-interview
Skill0 starsChanged 3 days agoApache-2.0View on GitHub
Conduct a collaborative product requirements interview between PM and engineering. Use when turning a brief or feature idea into a structured PRD and aspect-level specs through guided discovery and pressure-testing. Triggers on "prd interview", "requirements interview", "prd-interview".
In this repository10 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- assay-begin-fastSkill
- assay-beginSkill
- assay-end-fastSkill
- assay-endSkill
- assay-nextSkill
- assay-reportSkill
- assay-worktreesSkill
- prd-generatorSkill
- tech-planSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
14 rules. About 0 are common, 12 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈12
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 17
Do not create files until the Document phase.
- Distinctive: line 18
If the tool supports a read-only or plan mode, switch to it now.
- Very specific: line 26
Assay guardrail: before documenting, sanity-check the work against CLAUDE.md.
- Distinctive: line 34
Read the user's input — the issue dump, brief, or pasted requirements.
- Very specific: line 35
Read key files to understand the current architecture.
- Distinctive: line 38
Ask if the user is aware of prior art (NVIDIA's SkillSpector, OWASP's Agentic Skills Top 10) — offer to search if not.
- Distinctive: line 47
If the user has heard the challenge and wants to proceed, proceed.
- Distinctive: line 54
Requirements: Must-have vs. should-have vs.
- Distinctive: line 108
After writing, surface open questions and unresolved risks.
- Distinctive: line 115
Keep each spec focused and buildable by one engineer (or agent) at a time.
- Distinctive: line 123
Update existing PRD: Read the file, ask what changed, update in place.
- Distinctive: line 124
Existing PRD, no aspect specs yet: Run Aspect Decomposition Mode without re-running discovery.
- Distinctive: line 125
User says "just write it": Write from what you have, flag gaps in Open Questions, still include at least one challenge question.
- Distinctive: line 126
No brief exists: Run full discovery from conversation.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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
PRD Interview
Conduct a structured product requirements interview to turn a brief or feature idea into a complete PRD. This is a collaborative exercise — the PM hat brings product context, the engineering hat brings technical reality. Challenge assumptions. Pressure-test scope. Document what survives.
Do not create files until the Document phase. If the tool supports a read-only or plan mode, switch to it now.
Context
This skill is the first step in the brief-to-code pipeline (it's Phase 3 of assay-begin-fast).
Input is typically the gathered issue/brief dump at docs/planning/_card/issue.md plus the deep-dig understanding note.
Output is a structured PRD plus aspect-level specs that feed directly into tech-plan.
Assay guardrail: before documenting, sanity-check the work against CLAUDE.md. The moat is
the sandbox-observed behavior of a skill, diffed against what it declares, plus the compounding
findings corpus. It is not a static pattern scanner or a model's opinion of the skill's text
standing in for the verdict. If requirements drift toward a static-only signal deciding
MALICIOUS/CLEAN, flag it in the interview, not after.
Discover & Challenge
Read the user's input — the issue dump, brief, or pasted requirements.
Read key files to understand the current architecture. Assay is greenfield: today that means
docs/technical/CAPABILITY_ROADMAP.md (A1..A10), docs/technical/ARCHITECTURE.md, and
docs/ROADMAP.md, plus src/assay/ once it exists.
Ask if the user is aware of prior art (NVIDIA's SkillSpector, OWASP's Agentic Skills Top 10) —
offer to search if not.
Then pressure-test. Do not soften these.
- "What happens if we don't build this?"
- "Imagine this launched 6 months ago and failed. What went wrong?"
- "What are we choosing NOT to build if we build this?"
If the user has heard the challenge and wants to proceed, proceed.
Fill remaining gaps with focused questions, 2-3 at a time, grouped by topic:
- Users & Problem: Who has this problem? What's the cost of the status quo?
- Success: How will we measure it? Target numbers?
- Scope: What is explicitly out of scope?
- Requirements: Must-have vs. should-have vs. nice-to-have?
- Technical Fit: Which capability (
A1..A10) and phase does this belong to, and are its dependencies built? Where does it sit in the ingest → sandbox → capture → diff → verdict → bundle pipeline? - Verdict impact: Does this keep the sandboxed, observed run as the only thing that
assigns
CLEAN/FLAGGED/MALICIOUS? Does the floor (the four always-on finding classes) stay intact? What's the explicitUNVERIFIEDpath, with a named cause, when the sandbox can't run the skill or the dangerous path never triggers?
Skip what you can infer. Challenge vague answers. Flag technical pitfalls you read — don't wait to be asked.
Stop when the problem is clear without guessing, success metrics are measurable, must-haves have testable criteria, out-of-scope is explicit, and major technical risks are identified.
Confirm
Summarize: the problem, proposed approach, scope, success criteria, risks, and unresolved
concerns. Ask the user to confirm before writing. Confirm the feature slug for the directory name
(e.g., skill-ingest, sandbox-capture). Do not name the slug <type>-<id>.
Document
Omit sections that don't apply.
Filename: prd.md
Location: docs/planning/{slug}/.
docs/planning/{slug}/
├── prd.md ← this skill's output
├── {aspect}/ ← one directory per aspect
│ ├── spec.md ← this skill's decomposition output
│ ├── plan_YYYYMMDD.md ← tech-plan output
│ └── ...
└── ...PRD structure
- Problem Statement: What problem are we solving? For whom? Evidence it's real.
- Goals & Success Metrics: What does success look like? How will it be measured?
- User Personas & Scenarios: Who uses this and in what context? (Assay ICP: skill authors, registries like ClawHub/skills.sh, and enterprises deploying agents who need to know a skill does only what it says.)
- Requirements: Core features and behaviors, prioritized must/should/nice.
- Technical Considerations: Architecture fit, dependencies, integration points. Call out
sandbox isolation and verdict impact explicitly, and name the capability (
A1..A10). - Risks & Open Questions: Reference the roadmap's
R1..R9where one applies. - Out of Scope: Explicitly excluded features or concerns.
Include when relevant: Finding Schema, Declared-Scope Grammar, Sandbox Config Contract, Bundle Contract, Non-Functional Requirements.
After writing, surface open questions and unresolved risks. Then offer to continue into aspect decomposition.
Aspect Decomposition Mode (same skill)
- Propose aspect candidates (typically 2-8), each with a one-line boundary.
- Confirm aspect names with the user (
kebab-case). - For each confirmed aspect, write or update
docs/planning/{slug}/{aspect}/spec.md. - Keep each spec focused and buildable by one engineer (or agent) at a time.
Each spec.md: problem slice and user outcome, in-scope requirements, out-of-scope boundaries,
acceptance criteria (testable — the repo is test-first), dependencies/sequencing, open
questions/risks specific to this aspect.
Edge Cases
- Update existing PRD: Read the file, ask what changed, update in place.
- Existing PRD, no aspect specs yet: Run Aspect Decomposition Mode without re-running discovery.
- User says "just write it": Write from what you have, flag gaps in Open Questions, still include at least one challenge question.
- No brief exists: Run full discovery from conversation.
- Greenfield with no code to read: Ground the technical section in the roadmap/design docs and say plainly that no implementation exists yet.
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.

