evaluation agents

langwatch/langwatch

Agent

Total unimplemented-tagged scenarios: 76 Classified: 76.

redhat-community-ai-tools/harness-eval

Agent

General helper.

redhat-community-ai-tools/harness-eval

Agent

Run lint checks and fix issues automatically.

August1314/Michael-Polanyi

Agent

You are an analyzer agent for the michael-polanyi skill. Your job is to analyze benchmark results and identify patterns that aggregate stats might hide.

August1314/Michael-Polanyi

Agent

Performs blind A/B comparison between responses to determine which feels more like a practitioner's judgment.

August1314/Michael-Polanyi

Agent

You are a grader agent for the michael-polanyi skill. Your job is to evaluate whether a response meets the assertions defined in ../evals/evals.json.

JuanMarchetto/agent-skills

Agent

You are the Synthesizer for the Architect evaluation system. You receive structured outputs from multiple evaluator agents and synthesize them into a unified evaluation with clear scores, actionable recommendations, and strategic direction.

JuanMarchetto/agent-skills

Agent

You are the Analyzer agent for Founder Mode — Phase 7: Post-Run Analysis.

JuanMarchetto/agent-skills

Agent

You are the Architecture Generator agent for Founder Mode — Phase 2.

sattyamjjain/proofloop

Agent

Read-only isolated agent that evaluates skill/agent execution quality.

golemfoundation/octant-council-builder

Agent

Research on-chain activity, deployments, and usage metrics.

golemfoundation/octant-council-builder

Agent

Analyze project website, documentation, and public communications.

golemfoundation/octant-council-builder

Agent

Synthesize all evaluations into a final council report with recommendation.