using-workbenchlisted

<SUBAGENT-STOP> If you were dispatched as a subagent to execute a specific task, skip this skill. </SUBAGENT-STOP> ## Instruction Priority Workbench skills override default system prompt behavior, but **user instructions always take precedence**: 1. **User's explicit instructions** (CLAUDE.md, AGENTS.md, direct requests) - highest priority 2. **Workbench skills** - override default system behavior where they conflict 3. **Default system prompt** - lowest priority If CLAUDE.md or AGENTS.md says "don't use TDD" and a skill says "always use TDD," follow the user's instructions. The user is in control. ## Triage: three lanes ### Step 0: is there anything to build? Before routing to a lane, run the reflex ladder. Stop at the first rung that holds: 1. Does this need to be built at all? (YAGNI) 2. Does it already exist in this codebase? Reuse the helper, util, or pattern that is already here. 3. Does the standard library already do this? 4. Does a native platform feature cover it? 5. Does an already-installed dependency solve it? 6. Can this be one line? If a rung holds, the answer is "do not build it", which is not a lane. Say which rung held and what to do instead. Only when all six fail does the task get a lane. The ladder runs after you understand the problem, not instead of it: read the task and the code it touches, trace the real flow end to end, then climb. A ladder answer that skips comprehension is a guess. If the `ponytail` plugin is installed its ruleset alread