Skip to content
OpenAgentSkill
Browsing is open · Catalog captured 2026-11-21 · Automatic updates and account services are paused.
Education

reflect-codex-workflows

Compare evidence from a current or completed project with installed skills and decide whether to reuse, revise, create, or not skill the lesson. Use for workflow reflection or proposed skill upgrades. Do not edit skills without explicit authorization.

WildPigKing · Static checks on file

Installation

Read the source and dependencies before running a command. Copying does not install or execute anything.

npx skills add WildPigKing/codex-project-skill-manager --skill reflect-codex-workflows

About this skill


name: reflect-codex-workflows description: Compare evidence from a current or completed project with installed skills and decide whether to reuse, revise, create, or not skill the lesson. Use for workflow reflection or proposed skill upgrades. Do not edit skills without explicit authorization.

Reflect Codex Workflows

Review the objective, constraints, artifacts, corrections, failures, and verification evidence. Extract only lessons that could change execution in another project.

Read the repository skills menu first. When the user requests cross-project comparison, read the relevant user-level skills menu and only the shortlisted user SKILL.md files; do not scan the broader user home. For each distinct reflection task, shortlist at most three close skills and compare:

  • trigger and task boundary;
  • required inputs and environment;
  • decision rules and ordering;
  • completion and validation criteria;
  • failure recovery;
  • reusable scripts or references.

Choose one result:

  • Reuse when an existing skill already covers the workflow.
  • Revise when the boundary is the same and evidence adds a durable rule or validation method.
  • Create when the trigger or completion contract is materially distinct.
  • Do not skill when the lesson is project-specific, obvious, or unsupported.

Report the evidence, closest skill, marginal benefit, proposed trigger, one realistic validation scenario, and details that should not be generalized.

Ask before editing. When authorized, preserve user intent, keep the skill concise, update the registry, regenerate the menu, validate format, and label untested behavior honestly.