workbench-goal-mode
Workbench Goal Mode
Use this skill when an issue contains /goal, GOAL_MODE: yes, asks for full
autonomous completion, or asks an agent to keep working until a concrete outcome
is achieved.
Goal Mode is an execution wrapper, not a permission override. It keeps the objective alive across turns, reruns, local fixes, and partial evidence, while preserving existing approval, safety, and review gates.
Activation
Before changing files or live resources, post a compact lock:
GOAL_LOCK:
objective:
owner:
non_goals:
More from fearvox/multica-ultimate-workbench
workbench-token-context-discipline
Compact context, cache-aware execution, scoped evidence reads, and role-specific skill attachment discipline.
7workbench-memory-synthesis
Durable memory, session summaries, decision logs, issue closeout synthesis, and compact handoffs.
7workbench-research
Evidence-first research, architecture discovery, vendor documentation review, benchmark reading, and unknown-system analysis.
7workbench-conductor
Two-ring orchestration, routing, issue and comment discipline, and role boundaries for the Multica Workbench.
7workbench-review-qa
PASS, FLAG, and BLOCK review discipline for workflow review, QA verification, release checks, and issue closeout.
7workbench-design-docs
Product design, technical design documents, user-facing copy, specs, diagrams, and handoff documentation.
7