{"id":51060069,"url":"https://github.com/patrick204nqh/skills","last_synced_at":"2026-06-23T01:01:32.520Z","repository":{"id":357394834,"uuid":"1236543296","full_name":"patrick204nqh/skills","owner":"patrick204nqh","description":"Agent skills that help humans actually understand what the agent is telling them.","archived":false,"fork":false,"pushed_at":"2026-05-12T15:00:57.000Z","size":14,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-05-12T16:39:44.299Z","etag":null,"topics":["agent-skills","ai-agents","claude-code"],"latest_commit_sha":null,"homepage":"","language":null,"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/patrick204nqh.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":"2026-05-12T10:48:54.000Z","updated_at":"2026-05-12T15:54:25.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/patrick204nqh/skills","commit_stats":null,"previous_names":["patrick204nqh/skills"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/patrick204nqh/skills","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/patrick204nqh%2Fskills","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/patrick204nqh%2Fskills/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/patrick204nqh%2Fskills/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/patrick204nqh%2Fskills/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/patrick204nqh","download_url":"https://codeload.github.com/patrick204nqh/skills/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/patrick204nqh%2Fskills/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34671045,"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":"2026-06-22T02:00:06.391Z","response_time":106,"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","ai-agents","claude-code"],"created_at":"2026-06-23T01:01:28.467Z","updated_at":"2026-06-23T01:01:32.486Z","avatar_url":"https://github.com/patrick204nqh.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# Skills\n\nYou ask the agent a simple question — *how does this thing work?* —\nand what comes back is technically correct, three paragraphs long, and\nslides off your brain like water off a window. You read it twice. You\nstill couldn't repeat it to a colleague.\n\nThe agent isn't wrong. It's just not *communicating*.\n\nThat gap — between knowing a topic and being able to say it like a\nhuman would — is what these skills are for. Small, opinionated\ninstruction files that change how the agent talks: one analogy not\nthree, the answer before the journey, short sentences, no \"at its\ncore…\" filler. The agent already knows the material. The skill just\ngets it to land.\n\nPortable across Claude Code, Codex, Cursor, Gemini CLI, and any other\n[Agent Skills](https://agentskills.io)-compatible tool. Each skill is\na folder with a `SKILL.md` file, loaded only when relevant — so you\ncan keep many on hand without bloating context.\n\n## The skills\n\nThree families. Pick the one that matches what your reader needs.\n\n### Audience — *who is reading this?*\n\n- **[eli5](./skills/audience/eli5/SKILL.md)** — for when you want to *get* it.\n  Plain language, one strong analogy. Triggers on \"ELI5\", \"in simple\n  terms\", \"plain English\", \"dumb it down\".\n- **[exec-summary](./skills/audience/exec-summary/SKILL.md)** — for when a busy\n  leader has 60 seconds. Leads with the decision, not the mechanism.\n  Triggers on \"exec summary\", \"TL;DR for my boss\", \"for leadership\",\n  \"one-pager\".\n- **[eli-engineer](./skills/audience/eli-engineer/SKILL.md)** — for when a peer\n  engineer wants precision over accessibility. Names the data\n  structure, the invariant, the trade-offs. No analogies. Triggers on\n  \"ELI-engineer\", \"technical explanation\", \"skip the analogy\".\n\n### Structure — *how is the argument shaped?*\n\n- **[bluf](./skills/structure/bluf/SKILL.md)** — Bottom Line Up Front. The\n  answer in the first sentence, supporting detail after. Triggers on\n  \"BLUF\", \"TL;DR\", \"lead with the answer\", \"get to the point\".\n- **[minto-pyramid](./skills/structure/minto-pyramid/SKILL.md)** — for\n  proposals, recommendations, and decision docs. Answer first, then\n  grouped reasons, then detail. Triggers on \"structure this proposal\",\n  \"Minto\", \"pyramid principle\", \"make this argument tighter\".\n- **[feynman](./skills/structure/feynman/SKILL.md)** — for checking your own\n  understanding. Explain it plainly, then identify the part where you\n  got fuzzy. Triggers on \"do I actually understand this\", \"Feynman\",\n  \"test my understanding\", \"what am I missing\".\n\n### Format — *what does the reader need to see?*\n\n- **[diagram-it](./skills/format/diagram-it/SKILL.md)** — for when a\n  relationship is the point and prose buries it. Picks the right diagram\n  kind (flow, sequence, hierarchy, state), keeps it under five nodes,\n  emits mermaid. Triggers on \"draw this\", \"diagram\", \"visualise\",\n  \"show me the flow\", \"picture this\".\n- **[compare-table](./skills/format/compare-table/SKILL.md)** — for when\n  the reader is choosing between 2–4 options and prose buries the\n  differences. Four columns max, five-word cells, takeaway above,\n  nuance below. Triggers on \"compare\", \"what's the difference\", \"X vs Y\",\n  \"should I use A or B\", \"side by side\".\n- **[slide-deck](./skills/format/slide-deck/SKILL.md)** — for when you're\n  presenting to a room. One idea per slide, headline = takeaway (not\n  topic), speaker notes carry the story, no filler slides. Triggers on\n  \"make slides\", \"deck\", \"presentation\", \"slidify this\", \"turn this\n  into slides\".\n\n*More coming, slowly.*\n\n## Install\n\n### Claude Code (recommended)\n\nInstall as a plugin to get every skill at once and pick up updates\nautomatically:\n\n    /plugin marketplace add patrick204nqh/skills\n    /plugin install patrick204nqh@skills\n\n### Manual (any Agent Skills-compatible tool)\n\nCopy the skill folder you want into one of these locations:\n\n    ~/.claude/skills/             # Personal (Claude Code)\n    .claude/skills/               # Project-local, shared via git\n    ~/.agents/skills/             # Codex / generic Agent Skills\n\nThen just talk to your agent normally — skills trigger on intent, not\nslash commands (though most can be invoked explicitly too).\n\n## Philosophy\n\nThe best explanation is the one a colleague can repeat an hour later.\nThe best summary is the one a leader can act on without scrolling. The\nbest diagram is the one you don't have to squint at. These skills are\nopinionated rules — one analogy not three, the answer before the\njourney, five nodes not fifteen — because vague guidance doesn't\nchange behaviour. Specific guidance does.\n\nEach skill stays in the same shape: a single instruction file that\nshapes *how the agent communicates* — in words, in structure, or in a\npicture. No tools, no rendering, no dependencies. The skill tells the\nagent what to produce; the host renders it. Match the medium to what\nthe reader needs.\n\n## About\n\nBuilt by [Patrick](https://github.com/patrick204nqh).\nSteal them, fork them, send a PR if you sharpen one.\n\n## License\n\nMIT\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpatrick204nqh%2Fskills","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fpatrick204nqh%2Fskills","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpatrick204nqh%2Fskills/lists"}