LogoClawIndex
CasesSkillsAbout
LogoClawIndex

echoes-loop-diagnosing-risks - Diagnose Feedback Risks in LLM Recommenders

Diagnose, simulate, and mitigate bias amplification, hallucination propagation, and exposure polarization in LLM-powered recommender feedback loops.

Tags

Updated: 2026-10-16

Capabilities

Typical Inputs

Typical Outputs

What this skill does

  • Map LLM pipeline roles
  • Identify role-specific risks
  • Measure content-level risks
  • Measure ranking-level risks
  • Simulate recommendation cycles
  • Track ecosystem-level metrics
  • Trace amplification root causes
  • Apply targeted mitigations
  • Establish continuous monitoring

Inputs

  • Recommender pipeline architecture
  • LLM inputs and outputs
  • Ground-truth item catalog
  • User interaction histories
  • Ranking and exposure logs
  • User and item group metadata
  • Simulation cycle parameters

Outputs

  • Risk diagnostic report
  • Feedback-loop simulation harness
  • Content-level risk metrics
  • Ranking-level risk metrics
  • Ecosystem-level trend plots
  • Root-cause attribution results
  • Mitigation recommendations
  • Production monitoring alerts

Requirements

    Source

    • Spec: SKILL.md

    ClawIndex

    OpenClaw Skills & Use Case Index

    ClawIndex is an ecosystem-driven index of OpenClaw skills and real-world use cases.

    Index

    Skills·
    Cases

    Meta

    About·
    Disclaimer·
    Email·
    GitHub
    © 2026 ClawIndex All Rights Reserved.
    recommender systems
    large language models
    feedback loops
    bias amplification
    hallucination detection
    fairness monitoring
    risk diagnostics
    echo chambers
    Map LLM pipeline roles
    Identify role-specific risks
    Measure content-level risks
    Measure ranking-level risks
    Recommender pipeline architecture
    LLM inputs and outputs
    Ground-truth item catalog
    Risk diagnostic report
    Feedback-loop simulation harness
    Content-level risk metrics