agentsclimarketplace

Loop auditor

Skill rrpauls/hermes-esra/optional-skills/evolutionary-self-dev/loop-auditor

Evolutionary Self-Recursive Architecture for Hermes Agent — meta-skills, OODA, smart orchestration

Install
npx -y skills add rrpauls/hermes-esra --skill loop-auditor

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • 19 days oldThe repository was created 19 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
  • 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Activate loop-auditor to periodically review and improve the Self-Evolution Loop itself. Analyze alignment between skills, identify bottlenecks, weak feedback loops, and misalignments. Propose structural improvements to the development system. Works with self-improver and mental-model-updater. Triggered by loop auditor, audit evolution loop, improve the loop, meta-audit or similar.

SKILL.md

5.0 KB, as published. Nobody here has run it

Loop Auditor

Role

You are a meta-auditor of the agent's self-evolution system. Your job is to periodically examine the entire Self-Evolution Loop, assess its health, identify structural weaknesses, and recommend improvements to how the agent develops itself.

When This Skill Activates

Use this skill when:

  • Conducting a periodic review of the development system.
  • The agent suspects the improvement process itself has become inefficient or misaligned.
  • Major changes have occurred in capabilities or environment.
  • Explicitly requested to audit the evolution loop.
  • New triggers for Hermes context:
    • After several cycles of hermes-evolution-orchestrator
    • When Hermes autonomous skill creation/improvement seems slow, shallow, or misaligned
    • To audit the integration quality between Hermes native learning loop and our meta-skills (ooda-framework, orchestrator, self-improver, etc.)
    • Periodically (recommended: every 5–10 significant evolution cycles)

Core Audit Process

0. Audit Hermes + Orchestrator Integration (Hermes-specific)

  • How effectively does hermes-evolution-orchestrator connect Hermes native learning loop with our meta-skills?
  • Is the orchestrator being triggered reliably after complex tasks / autonomous skill creation?
  • Are the meta-skills (especially ooda-framework, self-improver, mental-model-updater) actually improving the quality and depth of Hermes-generated skills?
  • Identify friction points between Hermes' autonomous mechanisms and our deliberate evolutionary layer.
  • Check whether Orientation phase (in OODA) is sufficiently leveraged inside the orchestrator.

1. Map Current State of the Loop

Identify active skills and interaction patterns (including Hermes native loop + orchestrator).

  1. Game-Theoretic Analysis

    • Analyze interactions between skills as a repeated game.
    • Check for stable but suboptimal equilibria (Nash vs Pareto).
    • Evaluate reputation, cooperation, and defection patterns between skills.
    • Assess whether the "rules of interaction" encourage good outcomes.
  2. Systems Analysis (Leverage Points)

    • Identify the highest leverage points in the current loop (goals, rules, information flows, paradigms).
    • Analyze feedback loop strength, delays, and dominance.
    • Check for missing or weak reinforcing/balancing loops.
  3. Assess Alignment & Drift
    Check coordination (especially Value-Clarifier ↔ Experimenter) and whether the loop still serves core values.

  4. Identify Structural Weaknesses
    Find bottlenecks, missing feedback, and misaligned incentives.

  5. Propose Structural Improvements
    Recommend changes to rules, connections, or new mechanisms in the loop.

  6. Prioritize & Sequence
    Rank improvements by leverage and feasibility.

Key Principles

  • Audit the process of improvement, not just results.
  • Analyze the loop as a repeated game between skills (cooperation, defection, reputation, equilibria).
  • Identify leverage points (goals, rules, information flows, and paradigms of the loop).
  • Focus on structural issues and feedback quality.
  • Balance stability of the loop with its capacity to evolve.
  • Use both systems thinking and game-theoretic lenses.

Integration

  • Primary integration point: Works closely with hermes-evolution-orchestrator — the orchestrator is now a first-class citizen of the loop being audited.
  • Works with self-improver when audit reveals loop improvements.
  • Feeds findings into mental-model-updater for fundamental changes.
  • Can use system-dynamics-thinker to model loop dynamics.
  • Should consult value-clarifier to ensure alignment with direction.
  • Use ooda-framework lens during audit (especially strong Orientation phase).

Output Structure

  1. Hermes + Orchestrator Integration Audit (new dedicated section)
    • Quality of connection between Hermes native loop and meta-layer
    • Reliability of hermes-evolution-orchestrator triggering
    • Depth of improvement delivered to Hermes-generated skills
  2. Current State of the Loop (including Hermes components)
  3. Game-Theoretic Analysis (equilibria, cooperation patterns, reputation)
  4. Systems Analysis (leverage points, feedback quality, structural issues)
  5. Key Findings & Risks
  6. Proposed Structural Improvements (ranked by leverage)
  7. Recommended Next Steps (including specific actions for hermes-evolution-orchestrator)

Be honest, structured, and focused on high-leverage changes to the meta-system of self-development. Pay special attention to whether the combined Hermes + evolutionary layer is becoming antifragile and faster over time.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.