Chat + tool-call storytelling
Same coverage_query tool, two implementation tasks:
A asks about WebHost.Systems auth (coverage 0.486, 5 knowledge gaps)
and gets a learn recommendation; B asks about Graphonomous
MCP deployment (coverage 0.460) and also needs more knowledge.
The act threshold requires coverage ≥0.720 and uncertainty ≤0.350 —
both tasks fall short. A hypothetical review_goal shows what
happens when the gate opens.