Feature Spec Pipeline
Product Manager Official v1.1.30Generate detailed feature specifications from rough ideas, including user stories, acceptance criteria, and edge cases
by skrptiq
Sign in to import this skrpt into your workspace.
Sign inWhat's included 14 nodes
Feature Spec Pipeline
End-to-end workflow for generating detailed feature specifications from rough ideas through user stories, acceptance criteria, and edge case analysis
Acceptance Criteria Writing
Writing precise, testable acceptance criteria in Given/When/Then format with measurable outcomes
Edge Case Analysis
Identifying edge cases, failure modes, and boundary conditions through systematic analysis across six dimensions
Feature Decomposition
Breaking down high-level features into implementable components with clear boundaries and dependencies
Feature Spec Assembler
Compiling the feature brief, user stories, acceptance criteria, edge cases, and risk assessment into a single, cohesive feature specification document
User Story Writer
Converting decomposed feature components into well-structured, independently deliverable user stories with MoSCoW prioritisation
Acceptance Criteria Generator
Generate precise, testable acceptance criteria in Given/When/Then format for each user story
Edge Case Finder
Systematically identify edge cases across six analysis dimensions for a feature
Feature Brief Generator
Generate a structured feature brief from a rough idea, decomposing it into implementable components
Feature Spec Assembler
Assemble all pipeline outputs into a final, cohesive feature specification document
User Story Writer
Convert decomposed feature components into well-structured user stories with MoSCoW prioritization
Feature Specification Standards
Reference standards and industry best practices for feature specification writing
Feature Specification Guide
Guide to writing effective feature specifications using this pipeline
Feature Specification Template
Template for the final feature specification document produced by this pipeline
Requirements 1 service, 2 data types
- LLM Service
- pii
- product requirements
Security Passed
Detected
- Services: llm-service
- Permissions:
- Data Handling: pii
Version history 18 releases
GH#863 Wave 1 (#858-A1 class) — fix the intent/output mismatch (K-045). The bundle shipped `user-story-writer` and `feature-spec-assembler` prompts, and the docs said to run them, but the `execution:` block never did — so the workflow emitted polished intermediate analysis (stakeholder + risk) instead of the promised feature specification. Restored the intended bound graph: added a **User Story Writer** step and a **Feature Spec Assembler** step (each with a backing skill so they are `from_step`-addressable — `user-story-writing`, `feature-spec-assembly`), ordered `feature-decomposition → user-story-writing → acceptance-criteria-writing → edge-case-analysis → stakeholder-analysis → risk-assessment → feature-spec-assembly`, placing the assembler as the last content step before `language-polish`. Rewired every cross-step input to an **explicit `from_step` binding** via `context_params`/`{{step.context.*}}` instead of positional or mis-pointed `{{steps.<Title>.output}}` refs — user stories now feed acceptance criteria, edge cases, and the assembler; the assembler consumes the feature brief, user stories, acceptance criteria, edge cases, and risk assessment. Completed the tail: repinned `polish-language` 1.0.1→1.0.6 (the version exposing the bindable `source` slot) and bound `language-polish`'s `source` ← the assembled feature spec, so the `output_step` polishes the actual specification rather than its positional previous. Added 2 skills (skills 3→5, total 12→14). No new required inputs.
GH#845 — republish with American English (en-US) content, completing the source-only GH#805 flip that never reached the Hub. Copy only — no functional or behaviour change.
GH#745 — declare per-step `output: {name, type}` on every execution step (feature_brief/text, acceptance_criteria/list, edge_cases/list, stakeholder_analysis/text, polished_spec/text, compliance_verdict/decision, consistency_verdict/decision, input_gaps/decision, risk_assessment/text). Lights up the #744 rich flow-map with named, typed outputs. **Also corrects the input-gap-check step to its intended `validation` type** — its `step_type` was mis-indented (outside the parallel item) and dropped at parse time, so the step previously ran untyped; it is now a validation gate. Content + structural fix (GH#748); no bindings changes.
GH#645 Row 3b — migrate to K-037 dep-referenced schema. Strip 14 inline shared-content files and declare 14 hub-shared deps (UUID id + slug name + version + checksum from `gen-dep-checksums.mjs`). Closes pre-Step-3 inline-vendoring for this bundle.
Wave 2: re-signed with canonical engine signing pipeline.
Signature fix — RELEASE_NOTES.md now included in integrity checksum.
Initial catalogue release with full structural and content-quality validation. All scanner checks pass.
Release notes unavailable for this version.
Release notes unavailable for this version.
Release notes unavailable for this version.
Release notes unavailable for this version.
Release notes unavailable for this version.
Release notes unavailable for this version.
Release notes unavailable for this version.
Release notes unavailable for this version.
Release notes unavailable for this version.
Release notes unavailable for this version.
Release notes unavailable for this version.
More from Product Manager
View all →Batch Competitor Analysis
Analyze a list of competitors individually, then consolidate into a comparative matrix and strategic recommendations
Roadmap Generation Flow
Full product roadmap generation from vision through to themed, time-bound initiatives with dependency mapping, resource allocation, and stakeholder-ready output
User Research Pipeline
Plan and synthesize user interviews, surveys, and feedback into actionable insights