Delegate independent tasks to specialized agents working concurrently with isolated context.
Skill: experiment-retrospective
experiment-retrospective
Turn experiment reports and notes into summaries and new/updated result skills.
SKILL.md
Full skill instructions
Skill: experiment-retrospective
When to use
Use this when:
- The user starts with
<retrospective>, or - Requests a summary/lessons-learned across one or more experiments.
Behavior
-
Select inputs
- Use the user's description to identify relevant:
training_reports/*.md- Sections of
references/experiment-log.md
- If ambiguous, list candidate reports and ask the user to choose.
- Use the user's description to identify relevant:
-
Summarize experiments
- For each report:
- Extract: setup, hyperparameters, metrics, successes, failures.
- Write a markdown summary with:
- "What we tried"
- "Key findings"
- "What failed"
- "Open questions"
- For each report:
-
Update troubleshooting (if needed)
- If the experiments reveal new error patterns and their fixes:
- Propose new rows for
references/troubleshooting.md. - Ask the user for confirmation before editing the file.
- Propose new rows for
- If the experiments reveal new error patterns and their fixes:
-
Propose or update skills
- Decide what result skills should reflect these findings.
- For each skill:
- If new: start from
templates/skills/result-skill-template.md. - If existing: identify which sections to update.
- If new: start from
- Draft SKILL.md content that includes:
- General description, context, results summary, recommended practice, failure modes, and explicit guidance for future Codex usage.
- Use the user's description to fill in names, summaries, and assumptions, then remove any template sections that do not apply.
-
Ask before writing
- Present the proposed skill changes.
- Only create or modify files under
.codex/skills/with user approval.
-
Optional
- Append a summarized retrospective entry to
references/experiment-log.md(e.g. "2025-12-26 – Retrospective on IMDb+Amazon mixture experiments"). - Include a short "General description" line when logging.
- Append a summarized retrospective entry to
More Other skills
brand-guidelines
Applies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.
theme-factory
Toolkit for styling artifacts with a theme. These artifacts can be slides, docs, reportings, HTML landing pages, etc. There are 10 pre-set themes with colors/fonts that you can apply to any artifact that has been creating, or can generate a new theme on-the-fly.
internal-comms
A set of resources to help me write all kinds of internal communications, using the formats that my company likes to use. Claude should use this skill whenever asked to write some sort of internal communications (status reports, leadership updates, 3P updates, company newsletters, FAQs, incident reports, project updates, etc.).
Fundamental design principles for writing maintainable, testable Python code.
responsive-design
Modern responsive layouts using container queries, fluid typography, CSS Grid, and mobile-first strategies.
Generate Playwright tests from scenarios using interactive browser exploration and validation.
Automated website exploration and test case generation using Playwright.
Modern React state management with Redux Toolkit, Zustand, Jotai, and React Query for every state category.
gdpr-data-handling
GDPR-compliant data handling with consent management, data subject rights, and privacy controls.
Python error handling patterns including input validation, exception hierarchies, and partial failure handling. Use when implementing validation logic, designing exception strategies, handling batch processing failures, or building robust APIs.
refactor-plan
Detailed multi-file refactor planning with safety checks and rollback guidance.
