Agent skill

recursive-refiner

Self-improvement engine. Implements generate-critique-iterate loops for enhanced reasoning. Use when working through complex problems, synthesizing across domains, or when initial output needs refinement. Integrates with ego-check to prevent runaway confidence. Maximum 3 iterations to conserve tokens.

majiayu000534★ · 1 repos on radarProfile →
claude-codeMIT
Install
npx skills add majiayu000/claude-skill-registry --skill recursive-refiner-agentgptsmith-monadframework-06e14d65 --agent claude-code

Same command for any agent — swap --agent for codex, cursor, copilot.

Facts
Files in the skill folder: 2
SKILL.md size: 5 KB
Bundled scripts: none
Requires: - gremlin-brain-v2 - cognitive-variability - meta-pattern-recognition - chaos-gremlin-v2
Path: skills/ai-llm/recursive-refiner-agentgptsmith-monadframework-06e14d65/SKILL.md
Open the folder on GitHub →
Where it comes from
Stars: 534
Language: HTML

Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.

From the SKILL.md

# Recursive Refiner The engine for getting better at getting better. *Part of the GONAD: Gremlin Obnoxious Network of Actual Discovery* ## Purpose Implement controlled self-improvement loops: Generate → Critique → Iterate With governor (ego-check) to prevent confident nonsense. ## When to Use - Complex synthesis requiring multiple passes - Derivations that need verification - Outputs that feel incomplete on first attempt - When "good enough" isn't good enough - Framework development requiring iterative refinement ## The Core Loop ### Iteration 0: Generate Produce initial output. Don't overthink — get something down. ### Iteration 1-3: Critique & Refine For each iteration: **Step 1: Critique** Evaluate current output on: | Criterion | Question | Score 1-10 | |-----------|----------|------------| | Coherence | Does it hold together logically? | | | Grounding | Is it connected to verifiable facts/prior work? | | | Completeness | Are there obvious gaps? | | | Novelty | Does it add something, or just repackage? | | | Clarity | Could someone else follow this? | | **Step 2: Identify Weaknesses** - What's the weakest part? - What assumption is most questionable? - Where would Matthew push

What's inside
Steps it walks through
  1. Purpose
  2. When to Use
  3. The Core Loop
  4. Iteration 0: Generate
  5. Iteration 1-3: Critique & Refine
  6. Termination Conditions
  7. Integration with Other Skills
  8. Iteration Template
  9. Worked Example: Theoretical Synthesis
  10. Failure Modes
  11. Loop-lock
  12. Confidence Creep
  13. Polish Over Substance
  14. Complexity Creep
Ships with 1 file
  • metadata.json
More from claude-skill-registry
All skills →
About this skill
What does the recursive-refiner skill do?

Self-improvement engine. Implements generate-critique-iterate loops for enhanced reasoning. Use when working through complex problems, synthesizing across domains, or when initial output needs refinement. Integrates with ego-check to prevent runaway confidence. Maximum 3 iterations to conserve tokens.

How do I install it?

Run `npx skills add majiayu000/claude-skill-registry --skill recursive-refiner-agentgptsmith-monadframework-06e14d65 --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.

Where does this skill come from?

From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.

Is a popular skill a good skill?

Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.

Keep going