Adversarial coverage pass using exploratory testing heuristics to find untested cases.
STEP 3 — Generates the feature code. Run when the tasks are ready.
Validates requirements quality before implementation starts.
Integrates team feedback into the technical plan.
Integrates team feedback into the spec.
Matches all tools
Hooks run on every tool call, not just specific ones
Executes bash commands
Hook triggers when Bash tool is used
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Modifies files
Hook triggers on file write and edit operations
Modifies files
Hook triggers on file write and edit operations
Claude Code plugin for spec-driven development. Enables PMs, designers, and developers to collaborate in a structured process: spec → plan → code → deploy, with Claude assisting at every step.
1. Install the plugin
Open your terminal, type claude and press Enter. Once it loads, run:
/plugin marketplace add [email protected]:deuveme/product-flow.git
Then, in the same session, run:
/plugin install product-flow@product-flow
Close the terminal. If you had Claude Code Desktop open, close it and reopen it.
You are ready to use the plugin commands!
2. Update the plugin
Open your terminal, type claude and press Enter. Once it loads, run:
/plugin update product-flow@product-flow
Close the terminal and reopen Claude Code.
Commands:
/product-flow:start — start a new feature/product-flow:continue — advance to next step (repeatable)/product-flow:build — generate code (when plan is ready)/product-flow:submit — share code for review/product-flow:fix — fix issues found during testing or code review (TDD guaranteed)/product-flow:deploy-to-stage — publish to staging/product-flow:status — see where you are/product-flow:context — see memory usageWorkflow:
/product-flow:start → /product-flow:continue (repeat) → /product-flow:build → /product-flow:submit → /product-flow:deploy-to-stage
(DRAFT PR) (spec → plan, team feedback) (code, TDD) (exit DRAFT, review) (approved → merge)
↕ issues found?
/product-flow:fix
(TDD fix cycle)
| Document | For | Content |
|---|---|---|
docs/onboarding.md | PMs & designers | How to use the commands, full workflow, FAQs |
docs/guide.md | Dev team | Architecture, setup, how to modify skills |
docs/constitution.md | Tech lead & AI | Project governance, standards, scope discipline |
speckit.verify) that validates code against spec, plan, tasks, and constitution before submitting for review.speckit.verify-tasks checks that tasks marked done have real code behind them — not stubs or TODOs — using a 5-layer cascade (file existence, git diff, pattern matching, dead-code detection, semantic assessment).speckit.reconcile surgically updates spec, plan, and tasks when implementation diverges from the original design.research.md and decisions.md and proposes which decisions are worth promoting to project-level Architecture Decision Records. Proposals appear in the PR body under ## For Developers so the team can review them alongside the code.docs/adr/NNNN-<slug>.md in standard format (Context / Decision / Consequences) and committed to main alongside the merge.speckit.split analyzes a spec for scope creep using scope, size, and language signals. If a split is warranted, it trims the current spec and creates a new branch with a draft PR for the extracted feature — with vertical-slice validation and expand-contract warnings when features share entities.npx claudepluginhub deuveme/product-flow --plugin product-flowSpec-driven development for big features. When features get too big, plan mode gets too vague—leading to hallucinations during implementation. ShipSpec replaces vague plans with structured PRDs, technical designs, and ordered tasks that keep Claude grounded.
Provider-agnostic skills for autonomous product development: spec, task, implement, test, review, and summarize changes.
Specification-driven development workflow: specify → plan → tasks → implement
Helder's personal SDLC toolbelt for AI coding agents — from PRD to ship. Bundles the tracer-bullet workflow alongside TDD, code review, audits, and shipping skills.
End-to-end development workflow: design → draft-plan → orchestrate → review → pr-create → pr-review → pr-merge
HarnessFlow — From idea to shipped product: high-quality engineering workflows for AI agents. Spec-anchored SDD, gated TDD, evidence-based routing, independent reviews, and formal closeout.