{"id":50567883,"url":"https://github.com/analyticalmonk/explain-this","last_synced_at":"2026-06-16T16:30:56.146Z","repository":{"id":356519517,"uuid":"1228631114","full_name":"analyticalmonk/explain-this","owner":"analyticalmonk","description":"An agentic skills framework for creating interactive explainers for a topic, documents or codebase","archived":false,"fork":false,"pushed_at":"2026-06-01T11:43:55.000Z","size":242,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-06-04T16:23:58.773Z","etag":null,"topics":["agent-skills","claude-code","claude-code-plugin","codex","skills","skills-sh"],"latest_commit_sha":null,"homepage":"","language":"HTML","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/analyticalmonk.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-05-04T08:05:15.000Z","updated_at":"2026-06-01T11:43:58.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/analyticalmonk/explain-this","commit_stats":null,"previous_names":["analyticalmonk/explainer-skill"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/analyticalmonk/explain-this","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/analyticalmonk%2Fexplain-this","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/analyticalmonk%2Fexplain-this/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/analyticalmonk%2Fexplain-this/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/analyticalmonk%2Fexplain-this/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/analyticalmonk","download_url":"https://codeload.github.com/analyticalmonk/explain-this/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/analyticalmonk%2Fexplain-this/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34415240,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-16T02:00:06.860Z","response_time":126,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["agent-skills","claude-code","claude-code-plugin","codex","skills","skills-sh"],"created_at":"2026-06-04T16:01:09.561Z","updated_at":"2026-06-16T16:30:56.141Z","avatar_url":"https://github.com/analyticalmonk.png","language":"HTML","funding_links":[],"categories":[],"sub_categories":[],"readme":"# explain-this\n\nA Claude Code plugin for creating distill-style interactive explainers - single self-contained `index.html` pages with a sticky two-column layout, hand-built Canvas figures, and conversational prose, like the articles at [distill.pub](https://distill.pub). Zero dependencies, no build step.\n\n## The skills\n\n- **creating-explainers** - the hub. Turns a paper, blog post, transcript, or research report into an explainer, or researches a topic from scratch, or both. Owns the template, figures, voice, and the staged workflow.\n- **explaining-codebases** - explains a repository or set of source files: an onboarding overview of how a project is structured, or a deep-dive on how one mechanism works. Same output format, with code navigation and code-specific figures.\n- **fact-checking-explainers** - a gate the other two pass before delivery. Every checkable claim must trace to a source (or, for code, the real implementation), or be corrected or cut. No incorrect claim ships.\n\nEach skill triggers automatically when you describe the matching goal. You can also invoke fact-checking on its own.\n\n## Examples\n\nTwo interactive explainers built with this skill set. Open them and play with the figures - they step, drag, and toggle right in the browser.\n\n**[Superpowers: The Anatomy of an Agent Skill](https://www.akashtandon.in/interactive-explainers/superpowers/)**\n\n[![Superpowers explainer](docs/images/superpowers.png)](https://www.akashtandon.in/interactive-explainers/superpowers/)\n\n**[DSPy: Programming - Not Prompting - Language Models](https://www.akashtandon.in/interactive-explainers/dspy/)**\n\n[![DSPy explainer](docs/images/dspy.png)](https://www.akashtandon.in/interactive-explainers/dspy/)\n\n## Requirements\n\nNo API keys, no build tools. The skills use your agent's own file and web tools: research intake needs web search/fetch to be available, and the output is a single self-contained `index.html` you open in a browser.\n\n## Installation\n\n### Claude Code\n\nexplain-this is a Claude Code plugin, so it installs through the plugin marketplace. Adding the marketplace and installing the plugin pulls in all three skills at once:\n\n```\n/plugin marketplace add analyticalmonk/explain-this\n/plugin install explain-this@explain-this\n```\n\nThe first command points Claude Code at this repo on GitHub; the second installs the `explain-this` plugin from it. The skills land in `~/.claude/skills/` and trigger automatically.\n\nPrefer a local clone, or the repo is not published yet? Add the marketplace from a local path instead:\n\n```\ngit clone https://github.com/analyticalmonk/explain-this.git\n/plugin marketplace add ./explain-this\n/plugin install explain-this@explain-this\n```\n\n**Updating:** Claude Code does not auto-update plugins yet. To pull a newer version, refresh the marketplace and reinstall:\n\n```\n/plugin marketplace update explain-this\n/plugin install explain-this@explain-this\n```\n\n### Other agents that read Agent Skills\n\nThe repo also follows the open [Agent Skills](https://code.claude.com/docs/en/skills) layout (`skills/\u003cname\u003e/SKILL.md`). As of 2026 that format is read by several other coding agents, including **OpenAI Codex CLI**, **Google Gemini CLI**, and **GitHub Copilot** (in VS Code), as well as **Cursor** (which needs the skill placed manually) and tools like Cline, Windsurf, and Zed.\n\nThese agents do not use the Claude Code plugin marketplace. Install is manual: clone the repo and point your agent at the three skill folders, or copy them into whatever directory your agent loads skills from.\n\n```bash\ngit clone https://github.com/analyticalmonk/explain-this.git\n#   skills/creating-explainers/\n#   skills/explaining-codebases/\n#   skills/fact-checking-explainers/\n```\n\nCaveat worth knowing: these skills were authored and tested in Claude Code. They reference Claude Code tool names (Read, Edit, Bash, WebSearch / WebFetch) and invoke one another by name, so they will load in other Agent-Skills-compatible agents but are not tested there and may need light adaptation to your agent's tool set and skill-invocation syntax.\n\n### Not supported\n\nThere is no install for environments without a skills mechanism: the web chat apps (claude.ai, ChatGPT, and Gemini in the browser) and the bare model APIs. They cannot load `SKILL.md` skills at all. The only way to use the workflow there is to paste a skill's contents into the conversation by hand, which is not really supported and loses the lazy-loaded references that keep the skills light.\n\n## Usage\n\nYour agent picks the right skill up automatically when you say things like:\n\n- \"Make an interactive explainer about RLHF\" (research intake)\n- \"Turn this paper into a distill-style article\" (files intake)\n- \"Explain how this repo's scheduler works, as an interactive guide\" (codebase)\n\nWhichever path you take, the explainer is fact-checked before it is delivered: every claim is traced to its source or the real code, and anything unsupported is corrected or cut.\n\n## What's in this repo\n\n```\n.claude-plugin/\n  plugin.json                       # Claude Code plugin manifest\n  marketplace.json                  # marketplace entry for /plugin marketplace add\nskills/\n  creating-explainers/\n    SKILL.md\n    assets/article-template.html    # complete HTML skeleton, copy and fill in {{PLACEHOLDERS}}\n    references/                      # intake (files / research), figures, voice, template, palettes\n  explaining-codebases/\n    SKILL.md\n    references/                      # code intake, code-specific figure archetypes\n  fact-checking-explainers/\n    SKILL.md\n    references/verification-report-format.md\nevals/\n  evals.json                        # 5 reference prompts the skills are developed against\n```\n\nThe references load lazily - your agent reads them only when relevant, so the skills don't burn context up front.\n\n## License\n\nMIT - see [LICENSE](LICENSE).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fanalyticalmonk%2Fexplain-this","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fanalyticalmonk%2Fexplain-this","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fanalyticalmonk%2Fexplain-this/lists"}