Skip to content
Back to search

ELI5

SkillWorks with Claude CodeGitHub

ELI5 — A Claude Code skill that explains anything to anyone: kids, managers, engineers, parents. Adapts tone, vocabulary, and analogies to match the audience.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/dreambigou-eli5/badge.svg)](/listings/dreambigou-eli5)

HTML

<a href="/listings/dreambigou-eli5"><img src="/api/listings/dreambigou-eli5/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Copy the skill into your Claude Code skills directory:

git clone https://github.com/DreambigOu/ELI5.git
cp -r ELI5/skills/eli5 ~/.claude/skills/eli5

Then use it in Claude Code by saying things like "ELI5 this" or "explain this to my manager."

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 146 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # ELI5 — Explain Like I Am 5
  2. > A [Claude Code](https://claude.ai) skill that explains anything to anyone — kids, managers, engineers, parents. It adapts tone, vocabulary, analogies, and framing to match the audience.
  3. Ever needed to explain a technical concept to your manager? Or break down code for a 5th grader? ELI5 makes Claude automatically adjust its explanation style based on who's listening.
  4. Read the full blog post on how this skill was built: [Building an ELI5 Skill for Claude](https://andrewou.pages.dev/posts/building-an-eli5-skill-for-claude/)
  5. ## Supported Audiences
  6. | Category | Examples |
  7. |----------|---------|
  8. | **Ages** | 5, 10, 15, 20, 30, 40+ |
  9. | **Grade Levels** | 5th grade, Middle school, Senior High, College, Graduate school |
  10. | **Job Roles** | Manager, Engineer, Designer, Director, Product Manager |
  11. | **Relationships** | Wife, Husband, Parents, Kids, Friend |
  12. ## Usage Examples
  13. ```
  14. ELI5 what a database index is
  15. Explain this code to my manager
  16. Break down how git merge conflicts work for a 5th grader
  17. Explain this error to my mom
  18. Simplify this for a designer
  19. ```
  20. ## How It Works
  21. The skill detects the target audience from your prompt and calibrates:
  22. - **Vocabulary** — no jargon for kids, proper terminology for engineers
  23. - **Analogies** — toys and playground for age 5, business outcomes for managers
  24. - **Tone** — playful for children, professional for directors, warm for family
  25. - **Depth** — short and sweet for simple audiences, nuanced for grad students
  26. - **Framing** — impact/risk for managers, UX for designers, architecture for engineers
  27. ## Installation
  28. Copy the skill into your Claude Code skills directory:
  29. ```bash
  30. git clone https://github.com/DreambigOu/ELI5.git
  31. cp -r ELI5/skills/eli5 ~/.claude/skills/eli5
  32. ```
  33. Then use it in Claude Code by saying things like "ELI5 this" or "explain this to my manager."
  34. ## Evaluations
  35. Read the full guide on how the eval system works: [How to Evaluate a Claude Code Skill](https://andrewou.pages.dev/posts/how-to-evaluate-a-claude-code-skill/)
  36. ### Adding a New Test Case
  37. Test cases are defined in `eli5-workspace/evals.json`. Add a new entry to the `evals` array:
  38. ```json
  39. {
  40. "id": 3,
  41. "name": "explain-recursion-teenager",
  42. "prompt": "Explain recursion like I'm 15",
  43. "audience": "Age 15",
  44. "assertions": [
  45. "Uses social media, gaming, or phone references as analogies",
  46. "Tone is casual but not cringey — no 'fellow kids' energy",
  47. "Correctly explains the concept of a function calling itself",
  48. "Mentions a base case or stopping condition"
  49. ]
  50. }
  51. ```
  52. Each test case needs:
  53. - A **prompt** — what the user would say to Claude
  54. - A **name** — directory-friendly identifier for storing results
  55. - **Assertions** — specific, verifiable criteria to grade against (4 per test works well)
  56. ### Running Evaluations
  57. **Prerequisites:** [Claude Code CLI](https://docs.anthropic.com/en/docs/claude-code) installed and the skill installed at `~/.claude/skills/eli5/`.
  58. ```bash
  59. # Run all tests + auto-grade with pass/fail
  60. python eli5-workspace/run-evals.py
  61. # Run a single test
  62. python eli5-workspace/run-evals.py --test=1
  63. # Skip baseline, only test the skill
  64. python eli5-workspace/run-evals.py --with-skill-only
  65. # Grade existing outputs without re-running tests
  66. python eli5-workspace/run-evals.py --grade-only
  67. ```
  68. The script does three things:
  69. 1. **Runs each prompt** twice — once with the skill, once without (baseline)
  70. 2. **Auto-grades** every output against its assertions using Claude
  71. 3. **Prints a pass rate summary** comparing skill vs baseline
  72. Example output:
  73. ```
  74. --- Test 1: explain-db-index-age5 ---
  75. [with skill]
  76. PASS #1 — No technical jargon present
  77. PASS #2 — Uses book/page analogy and toy/messy room analogy
  78. PASS #3 — Sentences are short and conversational
  79. PASS #4 — Warm, enthusiastic tone with "huuuge", "super duper fast"
  80. [baseline]
  81. PASS #1 — No technical jargon found
  82. FAIL #2 — Uses phone book analogy, not child-friendly
  83. PASS #3 — Sentences are generally short
  84. FAIL #4 — Tone is informative but encyclopedic
  85. =========================================
  86. PASS RATE SUMMARY — Iteration 1
  87. =========================================
  88. With Skill: 10/12 passed (83.3%)
  89. Without Skill: 5/12 passed (41.6%)
  90. Delta: 41.7%
  91. =========================================
  92. ```
  93. Results are saved to `eli5-workspace/iteration-N/`, auto-incrementing with each run. Each test case produces `grading.txt` files with detailed evidence.
  94. ### Current Results
  95. See [eli5-workspace/eval-results.md](eli5-workspace/eval-results.md) for the full evaluation strategy and detailed grading.
  96. | Metric | With Skill | Without Skill | Delta |
  97. |--------|-----------|---------------|-------|
  98. | Pass Rate | **83.3%** | 41.6% | +41.7% |
  99. The biggest improvement is in audience-specific framing — especially for non-technical audiences like managers (0% baseline to 50% with skill).
  100. ## Contributing
  101. PRs welcome! Ideas for improvement:
  102. - Add more audience types (e.g., CEO, intern, journalist)
  103. - Add non-English language support
  104. - Improve evaluation coverage with more test cases
  105. ## License
  106. MIT

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.