"author:jiawei686" matching MCP servers:

Provides a single-purpose MCP tool that evaluates paper excerpts and returns calibrated ordinal scores for soundness, novelty, and clarity plus an accept/revise/reject decision with confidence metrics.

Enables code-review gating by converting a git diff into a typed verdict with safety probabilities, severity, and recommended actions for auto-merge or human review.

Enables agents to call typed code-review and content-moderation decision tools, returning structured verdicts, probabilities, and confidence-gated actions.

Enables agents to analyze contract excerpts and receive calibrated risk annotations across 10 default risk classes, including presence probabilities, overall risk levels, and human review recommendations.

Provides a single-purpose content-moderation gate that evaluates text and returns structured verdicts (spam/toxic probabilities, severity, confidence) with suggested actions (allow/review/block) for automated decision-making.