workflow

Plugin workflow from marketplace wiolett-industries in wiolett-industries/marketplace. Commands read from the repo's manifest — 21 days ago.

Skills (9)

• audit-flow

  Use for an explicit project, subsystem, diff, plan, quality, risk, security, architecture, maintainability, or readiness audit. Default to a quick local chat-only audit unless durable artifacts, several independent risk domains, or deep/exhaustive coverage are explicitly needed. The audit is read-on

• context-discovery

  Use after intent routing only when unresolved requirements, product intent, architecture, constraints, or tradeoffs can materially change scope, risk, acceptance criteria, or user-visible behavior. Inspect discoverable facts first and avoid exhaustive questioning.

• executing-plans

  Use to execute an approved active .workflow plan with milestone state updates, scoped task ownership, compaction recovery, selective worktree delegation, and one integration path. Do not use for direct fast-path work without a durable plan.

• finalizing-plan

  Use for standard or assurance completion, explicit independent review, risky handoff, or finalization of an active plan. Run one profile-appropriate verification/review path and stop at the completion latch. Fast work completes directly without this skill unless review is explicitly requested.

• intent-gate

  Use as a brief local routing step before non-trivial planning, execution, review, refactoring, debugging, or design when a shallow reading could change scope, risk, or user-visible behavior. Keep it silent when intent is clear; it does not imply an artifact or subagent.

• ui-contract

  Use as a supporting skill for substantial production UI and for any production UI change that names an existing/shared component, pattern, or analogous screen. Do not trigger for tiny copy/style changes without an architecture constraint, invisible frontend plumbing, or bounded mockups and throwaway

• using-workflow

  Use at conversation start, after context recovery, and for non-trivial engineering work to select exactly one primary workflow path plus one task-wide assurance and verification budget. Loading this router does not create artifacts, launch agents, or activate every workflow module.

• workflow-mcp

  Use only when creating, resuming, inspecting, updating, completing, or writing artifacts for an authorized .workflow plan or audit run through the bundled Workflow MCP tools. It is an operations contract, not a reason to create workflow artifacts.

• writing-plans

  Use when an authorized, sufficiently clear change needs a durable decision-complete plan under .workflow because execution is multi-step, risky, resumable, or explicitly requested. Do not create a plan for fast local work or read-only discussion.

Subagents (17)

• workflow_audit_prompt_writer

  Create standalone scoped audit prompts for large workflow audit runs.

• workflow_audit_reviewer

  Read-only domain auditor for workflow audit runs.

• workflow_audit_sanity_reviewer

  Optional single grouped sanity reviewer for deep/exhaustive audits with conflicting evidence, unsupported claims, duplicates, or severity inflation.

• workflow_combined_reviewer

  Optional single reviewer for a bounded standard-profile plan or change when a concrete review trigger exists; never required for the fast path.

• workflow_explorer

  Read-only repository explorer for broad multi-file or context-discovery sweeps. Use to locate files, map structure, trace how something works, and answer where/how questions without pulling large file contents into the orchestrator context. Returns a compact findings report, not file dumps.

• workflow_fix_triage

  Normalize several material review findings into scoped fix tasks without creating low-value review loops.

• workflow_implementer

  Implement one small, fully-specified, mechanical task in an isolated worktree (lightweight tier). Use when the approach is already decided and the edit is bounded.

• workflow_implementer_complex

  Implement one complex scoped task that requires architectural judgment and reasoning within a defined boundary (deep tier). Use for tightly-coupled or design-bearing changes where the approach may still need to be worked out.

• workflow_implementer_standard

  Implement one moderately complex scoped task that needs local reasoning within fixed boundaries (standard tier). Use when the approach is mostly decided but some judgment is required.

• workflow_intent_reviewer

  Rare read-only intent reviewer for unresolved high-assurance ambiguity where an independent interpretation can prevent a high-impact misread.

• workflow_master_auditor

  Compile substantial deep/exhaustive multi-review audits into master-audit.md, findings.json, and planning-input.md when synthesis justifies another agent.

• workflow_overall_reviewer

  Optional single primary reviewer for standard or assurance finalization when independent overall/code-quality review is justified.

• workflow_plan_overall_reviewer

  Optional assurance plan reviewer for completeness, task ordering, implementation realism, and integration; choose instead of stacking routine plan reviewers.

• workflow_plan_sanity_reviewer

  Optional assurance plan reviewer for missing context, unsafe assumptions, ambiguity, and execution blockers; choose instead of stacking routine plan reviewers.

• workflow_risk_reviewer

  Independent high-assurance reviewer for security, identity, destructive data changes, infrastructure, concurrency, public contracts, and broad blast radius.

• workflow_sanity_reviewer

  Optional primary reviewer for completed changes when integration breakage, unsafe assumptions, or contradictory verification creates a concrete review trigger.

• workflow_scope_compliance_reviewer

  Optional single primary reviewer when exact requirements, constraints, contracts, or non-goals create a concrete scope-compliance risk.

MCP servers (1)

• workflow

  runs: npx -y @wiolett/workflow@latest

Hooks (1)

• SessionStart

  node "${CLAUDE_PLUGIN_ROOT}/hooks/workflow-hook.cjs"

Similar plugins