learning

name

learning

description

Use at the end of a confirmed-successful problem-solving session where the path from problem to fix took more than one attempt and surfaced gotchas worth carrying forward. Trigger when the user invokes $learning, says "capture learnings", "what did we learn", "save learnings from this session", or after a debug arc finishes (passing tests, accepted fix, shipped feature) and the session contained failed attempts, contradicted assumptions, or revealed stale/missing entries in `.claude/reference/`.

Codex Adapter

Read ../../../.claude/skills/learning/SKILL.md completely and follow it as the canonical workflow for this skill.

• Resolve every relative script, reference, asset, and template path from the canonical skill directory under .claude/skills/learning/.
• Read ../../../.agents/CODEX-SKILL-COMPATIBILITY.md and apply the status and exceptions for this skill.
• Translate Claude Code tool names to the available Codex equivalent as described by the repository AGENTS.md.
• Before any Claude Task, Agent, or Workflow step, confirm that the current Codex session exposes the required multi-agent tools. Do not replace independent review with self-review.
• Skill invocation alone does not authorize commit, push, PR, merge, deploy, migration, dependency installation, or another outward action unless that action is the user's explicit request.
• Follow Codex system, developer, sandbox, approval, and tool instructions when they differ from the canonical workflow.
• Keep the canonical Claude skill unchanged unless the user explicitly asks to edit the workflow itself.