wb-flow Protocol: /wbWork Execution & Simulation Specification
This document defines the exhaustive behavior matrix for the /wbWork command. It serves as the definitive reference for deep task execution logic, wildcard ID selection, code generation constraints, and test-driven validation protocols.
1. Role & Definition Matrix
Role: The Implementer & Coder Target: Executes specific tasks from the active plan_*.md file. Core Protocol: Strict adherence to the Directed Acyclic Graph (DAG) defined in the plan. Cannot execute tasks out of chronological order if dependencies exist.
| Scenario | System Behavior |
|---|---|
| Target is UI Component | [PROCEED] Analyzes index.css or design system first. Ensures rich aesthetics, modern typography, and responsive layouts. |
| Target is Core Logic | [PROCEED] Employs Test-Driven Development (TDD). Analyzes existing tests and writes new assertions before modifying logic. |
| No Active Plan Found | [HALT] Protocol strictly forbids rogue coding. Emits "Cannot work without a blueprint" error and suggests /wbPlan. |
2. Argument & Criteria Resolution Matrix
The /wbWork command supports complex criteria targeting to allow precise or massive sweeping changes.
| Argument Type | Example | Parsing Logic | Simulated Output Profile |
|---|---|---|---|
| Single Task ID | Command: /wbWork -i="2" | Locks onto Task #2. Checks Dep column for DAG clearance. | Begins implementation of Task 2 exclusively. |
| Multi-Task Array | Command: /wbWork -i="1,3,4" | Parses comma-separated IDs. Sorts them chronologically. | Implements Task 1, then Task 3, then Task 4 sequentially. |
| Wildcard (All Tasks) | Command: /wbWork -i="*" | Extracts all tasks currently marked ⬜ Done. | Initiates a massive execution loop, processing every unblocked task in order. |
| Natural Language Selection | Command: /wbWork "do the UI tasks" | Parses the string against task descriptions. | Resolves fuzzy match to Task #3. Begins implementation. |
3. Flag Processing Matrix (Isolated Capabilities)
| Flag | Shortcut | Purpose | Example | Simulated Output Impact |
|---|---|---|---|---|
--id="<id>" | -i | Explicit task targeting (Supports singular, CSV arrays, and * wildcards). | Command: /wbWork -i="*" | [TASK] Wildcard detected. Queuing 4 pending tasks for sequential execution. |
--open | -o | Intercepts execution to ask the user for architectural guidance before coding. | Command: /wbWork -i="2" -o | [PROMPT] How should I handle the API timeout state before writing this component? |
--def | -d | Definitive mode. Executes code immediately using best-judgment without asking questions. | Command: /wbWork -i="2" -d | [DEF] Enforcing default error boundary implementation based on context.md. |
--can | -c | Pre-flight feasibility check. Evaluates if the task is actually codable given current context. | Command: /wbWork -i="1,2" -c | [CAN] Feasibility Check: Yes. Both tasks have sufficient API context to proceed. |
4. Omni-Channel Execution Pipeline (Flag Chaining)
/wbWork Exhaustive Simulation
The 'Massive Sweep' (-i="*" -d)
/wbWork -i="*" -d> /wbWork -i="*" -d[SYSTEM] Wildcard detected. 3 tasks queued.
[WORK] Implementing Task 1: JWT Handshake.
[CODE] Updating tierEnforcement.js...
[PLAN] Marking Task 1 as ✅ Done.
[WORK] Implementing Task 2...
⚠️ Note: User wants the agent to complete every single unblocked task in the plan immediately, without stopping to ask questions.
5. Operational Edge Cases & Protocol Faults
| Fault Trigger | System Detection | Resolution / Output |
|---|---|---|
| Blocked Dependency | User runs /wbWork -i="3", but Task 3 depends on Task 2 (which is ⬜). | ❌ Error: DAG Violation. Cannot execute Task 3 until Task 2 is marked ✅ Valid. |
| Wildcard DAG Halt | User runs -i="*". Task 1 fails compilation. | ⚠️ Warning: Task 1 failed. Halting wildcard queue to prevent cascading errors in Task 2 and 3. |
| Code Breakage | Syntax error in generated code chunk. | ⚠️ Warning: AST validation failed. Re-evaluating code implementation. |
| Invalid ID | User runs -i="99" (Task doesn't exist). | ❌ Error: Task ID 99 not found in active plan. |
← Home · Commands · Install | wb-flow on npm · flow.wbc-ui.com · wi-bg.com
⏱️ Matrix Task Duration Estimations
In the ## 🌊 Next Executable Sequence matrix table, each task dispatch cell appends the estimated task duration extracted from the task table's Est. (min) column, formatted as *(⏱️ <min> min)*:
markdown
`/wbWork plan.md --id=B23`<br>→ *DeepSeek V4 Pro* *(⏱️ 15 min)*💡 Pre-Flight Explanation Blueprint Gate (--as)
- Standard Mode (default, without
--as): Matrix cells contain ONLY the direct execution command:markdown`/wbWork plan.md --id=B23`<br>→ *DeepSeek V4 Pro* *(⏱️ 15 min)* - Explanation Mode (with
--as="<style>"): Matrix cells prepend/wbExplain:markdown`/wbExplain plan.md --id=B23 --as="expert,steps"`<br>`/wbWork plan.md --id=B23`<br>→ *DeepSeek V4 Pro* *(⏱️ 15 min)*
🎛️ Universal model flags (2026-08-01)
| Flag | Alias | Effect |
|---|---|---|
--planner= | -p | 🧠 Planner chain — persists via /wbModel |
--validator= | -v | ✅ Validator chain — persists |
--worker= | -w | 🔨 Worker chain — persists. ⚠️ -w is not --wave |
--mechanical= | -m | 📋 Mechanical chain — persists |
--model= | -M | Delegate this run. Highest priority: outranks role routing, the roster and the executor≠validator rule |
--wave=<L>:<R> | -W | Run one cell — :P Planner · :V Validator · :W Worker · :M Mechanical |
/wbWork Wave Execution
Executing a Wave Cell (--wave=A:W)
/wbWork plan.md --wave="A:W" -M=$WORKER> /wbWork plan.md --wave="A:W" -M=$WORKER[SYSTEM] Reading plan.md for Wave A, Worker cell...
[TRIAGE] Delegating to claude:opus 5 (overriding default roster).
[WORK] Implementing Task B23: Wave A Worker dispatch.
[OK] Task complete. ⏱️ 15 min elapsed.
[PLAN] Updating 🌊 Next Executable Sequence matrix...
Generated Output (
| Wave | 🧠 Planner | 🔨 Worker | ✅ Validator | 📋 Mechanical |
|---|---|---|---|---|
| Wave A | ✅ Done | ✅ Done (claude:opus 5) | ⬜ Pending | ⬜ Pending |
| Wave B | ⬜ Pending | ⬜ Pending | ⬜ Pending | ⬜ Pending |
⚠️ Note: The --wave flag lets you surgically execute a single cell in the plan's Next Executable Sequence matrix, delegating on the fly with -M.
Role flags are shorthand for running /wbModel first. An unknown role letter in --wave exits non-zero rather than silently running the whole row. Precedence: -M → role flag → plan-header roster → model_recommendations.md → defaults.
After the 🌊 matrix, a copy/paste block of bare runnable commands is printed — no table markup, no <br>, no duration annotations.