venture-challengelisted

# Venture Challenge Target venture directory: `$ARGUMENTS` Read the complete target venture state and current research. Read `framework/anti-bias.md`. Invoke the `challenger` subagent with the venture's current thesis and evidence set. Persist the synthesis to `research/YYYY-MM-DD-challenge.md`. Update `venture.yaml` only for newly explicit assumptions, risks, blocked unknowns, and canonical guardrails. Do not silently change earlier evidence. The output must include: - strongest failure modes; - unsupported leaps; - contradictory evidence; - critical unknowns; - explicit do-not-build items; - evidence that would change the challenge assessment. Any do-not-build item persisted to state must be an object with a stable `DNB###` ID, statement, and reason/evidence where available. Never use positional references. Replace `next_action` with the next stable `N###` object whose single instruction is to run the decision gate. Set `type: decision`; point `assumption_id` at the principal blocker when one clearly dominates, otherwise null.