prd-draft/SKILL.md
Continuation skill: assemble a PRD from validated artifacts, run quality gates, and publish. Runs after prd-interview when both problem-brief.md and decision-ledger.md exist. Not for writing PRDs from scratch.
npx skillsauth add sanurb/skills prd-draftInstall 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.
Assemble the PRD from validated artifacts, gate it for quality, and ship it. Every section is filled from the problem brief and decision ledger — if a section is empty, the upstream interview missed something.
Activation guard: This is a continuation skill. Only run when both
/tmp/prd-{name}/problem-brief.mdand/tmp/prd-{name}/decision-ledger.mdexist from prior pipeline runs. Do not activate independently.
Read both artifacts:
/tmp/prd-{name}/problem-brief.md/tmp/prd-{name}/decision-ledger.mdIf either is missing, stop. Tell the user which upstream skill to run first.
Fill the template from references/prd-template.md by mapping:
| Template Section | Source | |-----------------|--------| | Problem Statement | Problem brief → Pain + Current State + Impact | | Success Metrics | Decision ledger → B1 | | User Stories | Synthesize from problem brief + B2 + B3 | | Solution Approach + Key Decisions | Decision ledger → B3 | | Modules | Decision ledger → B7 | | Acceptance Criteria | Decision ledger → B5 + B1 | | Non-Functional Requirements | Decision ledger → B6 | | Scope / Non-Goals | Decision ledger → B2 | | Assumptions & Risks | Decision ledger → Assumptions + Deferred | | Open Questions | Decision ledger → Deferred Decisions | | Testing Strategy | Decision ledger → B7 | | Rollout & Measurement | Decision ledger → B8 |
Context adaptation — adjust weight based on the project:
| Context | Action | |---------|--------| | Startup / early-stage | Keep only: Problem, Metrics, User Stories, Approach, Scope, Open Questions | | Scale-up | Full template | | Enterprise / regulated | Full + add compliance traceability | | API / platform | Full + add Interface Contract section | | AI / agent product | Full + add Eval Framework, Guardrails sections |
Execute every check from references/quality-gates.md against the draft.
Present the draft to the user with gate results. Ask: "Does this capture our shared understanding? Anything missing, wrong, or over-specified?"
Iterate until the user approves.
Write the draft to /tmp/prd-{name}/prd-draft.md.
If the user wants a GitHub issue, execute:
gh issue create --title "{Feature Name} — PRD" --body-file /tmp/prd-{name}/prd-draft.md
After creation, tell the user the issue URL and suggest: "Run prd-to-issues to break this into vertical slice implementation issues."
The PRD is NOT ready to publish unless ALL of these are true:
/tmp/prd-{name}/prd-draft.md + optional GitHub issueprd-discover → prd-interview → this skill → prd-to-issuesdevelopment
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.