smithery.ai

repo-applying-maker-checker-fixer

Three-stage content quality workflow pattern (Maker creates, Checker validates, Fixer remediates) with detailed execution workflows. Use when working with content quality workflows, validation processes, audit reports, or implementing maker/checker/fixer agent roles.

First seen Mar 19, 2026

Installation

$ npx skills add https://smithery.ai

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from smithery.ai · top by installs.

npx skills add https://smithery.ai

Browse all from smithery.ai

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 5,229 B
  • docs SUMMARY.md 308 B

History

  1. First seen on skills.sh
  2. First recorded snapshot · 1 installs

SKILL.md

Maker-Checker-Fixer Pattern (Comprehensive)

Complete guidance on the three-stage content quality workflow pattern used across repository agent families for systematic content creation, validation, and remediation.

Purpose

Use when implementing content quality workflows, working with maker/checker/fixer agents, validating content against conventions, applying validated fixes from audit reports, or creating new checker/fixer agents.

The Three Stages

Maker creates/updates content with cascading dependency management. Checker validates content against conventions and generates a criticality-categorized audit report (non-destructive). Fixer re-validates each finding, then applies only HIGH-confidence fixes, skipping MEDIUM and FALSE_POSITIVE for manual review.

See [Stage 1: Maker and Stage 2 Checker Role](./reference/stage1-maker-and-checker-role.md) for the Maker workflow and the Checker's role/criticality categories.

The Checker's 5-step process (Initialize → Validate → Finalize) is detailed in [Checker Workflow: 5-Step Process](./reference/checker-workflow-steps.md) and [Checker Report Finalization and Progressive Writing](./reference/checker-finalization.md).

The Fixer's 6-step process is detailed across [Fixer Role, Priority, and Report Discovery](./reference/fixer-role-and-detection.md), [Fixer Mode Parameter Handling and Fix Application](./reference/fixer-mode-and-application.md) (lax/normal/strict/ocd), and [Fix Report Structure and Trust Model](./reference/fixer-reporting-and-trust.md) (why fixers lack web tools).

Lifecycle-Owned Validation

Before any quality-gate checker, fixer, or recheck invocation, apply the canonical [lifecycle ownership Step 0](../../../repo-governance/workflows/meta/workflow-identifier/check-fix-lifecycle-validation-ownership.md). Exclude registry-owned predicates from prompts and report them through lifecycle-status; keep final-status domain-only.

Common Workflows and Agent Families

See [Common Workflows and Agent Families](./reference/common-workflows-and-agent-families.md) for the basic and iterative maker→checker→fixer sequences and the repository-wide list of agent families using this pattern.

Best Practices and Common Mistakes

See [Best Practices and Common Mistakes](./reference/best-practices-and-common-mistakes.md) for role-specific DO/DON'T guidance and the most common mistakes for all roles, checkers, and fixers.

Tool Requirements

See [Tool Requirements](./reference/tool-requirements.md) for the typical tool set for checkers (read-only + report generation) and fixers (no web tools — trust checker's verification).

Preventing Iteration Loops

Without explicit safeguards, checker-fixer workflows can loop indefinitely. See [Preventing Iteration Loops](./reference/preventing-iteration-loops.md) for the four structural safeguards: FALSE_POSITIVE persistence, scoped re-validation, self-verification after bash edits, and escalation after repeated disagreements.

Integration with Conventions

The pattern integrates with:

  • [Criticality Levels Convention](../../../repo-governance/development/quality/criticality-levels.md) - Checkers categorize by criticality, fixers use for priority
  • [Fixer Confidence Levels Convention](../../../repo-governance/development/quality/fixer-confidence-levels.md) - Fixers assess confidence, combine with criticality
  • [Temporary Files Convention](../../../repo-governance/development/infra/temporary-files.md) - Checker/fixer reports stored in local-tmp/<agent-family>/
  • [Repository Validation Methodology](../../../repo-governance/development/quality/repository-validation.md) - Standard validation patterns
  • [AI Agents Convention](../../../repo-governance/development/agents/ai-agents.md) - Agent structure, tool permissions, color coding

References

  • [Maker-Checker-Fixer Pattern Convention](../../../repo-governance/development/pattern/maker-checker-fixer.md) - Complete pattern documentation
  • [Criticality Levels Convention](../../../repo-governance/development/quality/criticality-levels.md) - Severity classification
  • [Fixer Confidence Levels Convention](../../../repo-governance/development/quality/fixer-confidence-levels.md) - Confidence assessment
  • [Temporary Files Convention](../../../repo-governance/development/infra/temporary-files.md) - Report file organization and naming
  • [Repository Validation Methodology](../../../repo-governance/development/quality/repository-validation.md) - Validation patterns

Related Skills

  • repo-assessing-criticality-confidence - Deep dive into criticality/confidence levels and priority matrix
  • repo-generating-validation-reports - UUID chain generation, report format, progressive writing
  • repo-understanding-repository-architecture - Understanding the six-layer governance and where patterns fit