prd-interview/SKILL.md
Continuation skill: walk the design tree branch-by-branch until every decision is explicit. Runs after prd-discover when problem-brief.md exists. Not for standalone use, brainstorming, or solution design.
npx skillsauth add sanurb/skills prd-interviewInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Resolve every design decision through a dependency-ordered interview. One branch at a time. No child question asked while the parent is unresolved.
Activation guard: This is a continuation skill. Only run when
/tmp/prd-{name}/problem-brief.mdexists from a priorprd-discoverrun. Do not activate independently.
Read /tmp/prd-{name}/problem-brief.md. If missing, stop and tell the user to run prd-discover first.
From the problem brief, map decision branches in this fixed order:
| # | Branch | Depends On | Status | |---|--------|-----------|--------| | B1 | Success Criteria | — | open | | B2 | Scope Boundaries & Non-Goals | — | open | | B3 | Solution Approach | B1, B2 | blocked | | B4 | Data Model & State | B3 | blocked | | B5 | Edge Cases & Failure Modes | B3 | blocked | | B6 | Non-Functional Requirements | B2 | blocked | | B7 | Module Design & Testing | B3, B5 | blocked | | B8 | Rollout & Measurement | B1, B3 | blocked |
Show this map to the user before asking the first question.
For each branch (lowest unblocked first), ask the probe questions from references/interview-branches.md.
Per-decision protocol (every decision, no exceptions):
After resolving a branch, update its status to decided and unblock dependents.
When all branches are decided or deferred-with-owner, write to /tmp/prd-{name}/decision-ledger.md.
Copy the template from assets/decision-ledger-template.md and fill every section from the interview.
The decision ledger is NOT complete unless ALL of these are true:
decided or deferred-with-owner — no branch left opendeferred decision has an owner, a revisit trigger, and what it blocks/tmp/prd-{name}/decision-ledger.mdprd-draftdevelopment
Sets up an `## Agent skills` block in AGENTS.md/CLAUDE.md and `docs/agents/` so the engineering skills know this repo's issue tracker (GitHub, GitLab, fp, or local markdown), triage label vocabulary, and domain doc layout. Run before first use of `fp-plan`, `fp-implement`, `fp-review`, `to-issues`, `to-prd`, `triage`, `diagnose`, `tdd`, `improve-codebase-architecture`, or `zoom-out` — or if those skills appear to be missing context about the issue tracker, triage labels, or domain docs.
development
Build a throwaway prototype to flush out a design before committing to it. Routes between two branches — a runnable terminal app for state/business-logic questions, or several radically different UI variations toggleable from one route. Use when the user wants to prototype, sanity-check a data model or state machine, mock up a UI, explore design options, or says "prototype this", "let me play with it", "try a few designs".
tools
Control herdr (a terminal-native agent multiplexer) from inside it. Manage workspaces and tabs, split panes, spawn sibling agents, read pane output, and wait for state changes — all via CLI commands that talk to the running herdr instance over a local unix socket. Use when running inside herdr (HERDR_ENV=1). Do not use outside herdr.
documentation
Compact the current conversation into a handoff document for another agent to pick up.