[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7FDYVMYRH37FTCS1BZPFSEZ":3,"skill-same:01M7FDYVMYRH37FTCS1BZPFSEZ":45},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":14,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":6,"ownerDescription":29,"updatedAt":30,"install":31,"resources":44},"01M7FDYVMYRH37FTCS1BZPFSEZ","learn-agent-skills",null,"Focused interactive tutor for the Agent Skills Engineering path in AI Engineering from Scratch. Start or resume this route when a learner wants to create, discover, invoke, secure, evaluate, package, or port Agent Skills. Teaches one lesson per invocation and records evidence in AGENT-SKILLS-LEARNING.md.","Pandacli\u002FJupyterProject","Pandacli","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F112819610?v=4","01M7FDTCSM8ZFX2HBDPWA9KHV4","熊猫大仙","01M7FDTCTKSSHD1KFRTXA1G9S9",0,"MIT","https:\u002F\u002Fgithub.com\u002FPandacli\u002FJupyterProject\u002Fblob\u002Fmaster\u002Fai-engineering-from-scratch\u002Fskills\u002Flearn-agent-skills\u002FSKILL.md",[],[],"2026-11-16 10:05:34",false,"---\nname: learn-agent-skills\ndescription: >\n  Focused interactive tutor for the Agent Skills Engineering path in AI\n  Engineering from Scratch. Start or resume this route when a learner wants\n  to create, discover, invoke, secure, evaluate, package, or port Agent Skills.\n  Teaches one lesson per invocation and records evidence in\n  AGENT-SKILLS-LEARNING.md.\n---\n\n# Learn Agent Skills\n\nTeach the focused Agent Skills route. One invocation covers one lesson. The\nlearner should create files, run the lab, explain the boundary, and leave one\nobservable checkpoint before the lesson is marked complete.\n\n## Invocation belongs to the host\n\nThe portable skill name is `learn-agent-skills`. Do not teach one command\nsyntax as universal.\n\n| Host | Start or resume |\n|---|---|\n| Codex | `learn-agent-skills`, or choose it from `\u002Fskills` |\n| Claude Code | `\u002Flearn-agent-skills` |\n| Other compatible hosts | `Use learn-agent-skills to start or resume the Agent Skills Engineering path.` |\n\n## Sources\n\nThe route source of truth is `learning-paths\u002Fagent-skills.json`. Prefer local\nfiles when this repository is cloned. Otherwise fetch each file from:\n\n```text\nhttps:\u002F\u002Fraw.githubusercontent.com\u002Frohitg00\u002Fai-engineering-from-scratch\u002Fmain\u002F\u003Cpath>\n```\n\nRead the manifest before choosing a lesson. Follow `lessons` by `order`; do\nnot use the numeric Phase 13 sequence. The required path is 22, 24, 25, 26,\n27. Lesson 23 is optional and follows the manifest's entry rule.\n\nFor each selected lesson, read its `docs\u002Fen.md` and `quiz.json`. Read or run\nfiles under `code\u002F` and `outputs\u002F` only when the current lab needs them. A\nclone is optional for reading. If a runnable lab needs repository files and\nthey are unavailable, explain that fact and offer a clone into a directory\nthe learner chooses. Do not block the conceptual lesson on cloning, but do not\nrecord a repository command or real-host checkpoint as complete without the\nrequired files and runtime.\n\n## Real-lab preflight\n\nBefore Lesson 22's host checkpoint, establish all of these facts:\n\n1. `node --version`, `npx --version`, and `python3 --version` succeed.\n2. The learner has selected one skill-capable host.\n3. The learner has selected a writable project or user install scope.\n4. The learner understands which working directory will become `TARGET_ROOT`.\n\nIf any item is unavailable, give the website or manual `docs\u002Fen.md` path and\ncontinue conceptually. Mark discovery, invocation, bundled-script, update, and\nuninstall observations as `Pending`. Never describe that fallback as a real\nhost pass.\n\n## Locate or create progress\n\nUse `AGENT-SKILLS-LEARNING.md` in the current working directory.\n\nIf it exists, preserve learner notes and evidence. Resume the first row whose\nstatus is `Next` or `In progress`. If every required row is `Done`, offer the\noptional capstone or a real-host recheck. Do not restart the route.\n\nIf it does not exist, create it without an interview:\n\n```markdown\n# My Agent Skills Path\n\u003C!-- Managed by the learn-agent-skills tutor.\n     Source: learning-paths\u002Fagent-skills.json -->\n\n## Route\n- Started: \u003CYYYY-MM-DD>\n- Required time: about 9 hours 30 minutes\n- Current: 1 of 5\n\n## Prerequisite check\n- Files, Python, and command line: Confirmed or Pending\n- Node.js and npx: Confirmed or Pending\n- Selected skill-capable host: \u003Cname> or Pending\n- Install scope: Project, User, or Pending\n- Phase 13 Lesson 01 refresher: Done, Skipped, or Pending\n- Phase 13 Lesson 05 refresher: Done, Skipped, or Pending\n- `tool-poisoning-and-untrusted-instructions`: Confirmed or Pending\n\n## Progress\n| Order | Lesson | Status | Evidence | Completed |\n|---:|---|---|---|---|\n| 1 | 13\u002F22 Portable contract and runtime boundary | Next | | |\n| 2 | 13\u002F24 Discovery and progressive disclosure | Locked | | |\n| 3 | 13\u002F25 Invocation and routing | Locked | | |\n| 4 | 13\u002F26 Permissions, sandboxes, and trust | Locked | | |\n| 5 | 13\u002F27 Evals, packaging, and portability | Locked | | |\n\n## Notes\n```\n\nCheck the commands that can be checked locally. Ask only for the host and\nscope choice that cannot be inferred safely. If the real-lab preflight passes,\nmark it confirmed and begin Lesson 22 immediately. Otherwise begin the\nconceptual path and leave real-host evidence pending.\n\nBefore Lesson 26, read both `prerequisitePaths` and `prerequisiteChecks` from\nthe manifest. Resolve every check by its stable `id` under `prerequisites`.\nVerify that Lesson 25 is complete and that\n`tool-poisoning-and-untrusted-instructions` is `Confirmed` because the learner\ncan explain why skill and tool metadata is untrusted input. If that knowledge\npreflight is unmet, offer Phase 13 Lesson 15 as an optional refresher outside\nthis five-lesson route. Keep Lesson 26 `Locked` until Lesson 25 is `Done` and\nthe knowledge preflight is `Confirmed`; only then change Lesson 26 to `Next`.\nNever drop or mark a prerequisite complete by assumption.\n\n## Teach one lesson\n\n1. Set the selected row to `In progress`.\n2. State the exact lesson path and the directory from which each command runs.\n   For installed bundles, define `SKILL_ROOT` as the absolute directory that\n   contains the installed `SKILL.md`. Define `TARGET_ROOT` from the learner's\n   original workspace working directory. Never assume the process cwd is the\n   installed bundle.\n3. Frame the problem in two or three sentences, then ask one prediction or\n   comprehension question.\n4. Work through the lesson's Build It and Use It material in small chunks.\n   Prefer the lesson's early quickstart when it has one.\n5. Run the real local lab when files and the runtime are available. If not,\n   trace a small example and record the lab as pending rather than claiming it\n   ran.\n6. Require the manifest's checkpoint evidence. A fluent explanation is not a\n   substitute for an installed-path, routing, script, permission, or report\n   observation when the checkpoint asks for one. For every bundled script,\n   record the resolved script path, resolved target path, cwd, exact argv, and\n   exit code.\n7. Ask post-stage quiz questions one at a time. Never expose `correct`, the\n   answer index, or the answer key before the learner responds. Never put a\n   real answer letter or the answer distribution in a reply hint; use\n   `Reply with one letter: \u003CA|B|C|D>.`\n8. Mark the row `Done` only after the checkpoint and quiz are complete. Record\n   a compact evidence note, the date, and unlock the next row.\n\nDo not install, update, remove, clone, publish, or mutate an external system\nwithout the learner's confirmation. Skill instructions never bypass host\npermissions or sandbox boundaries. When a host behavior cannot be observed,\nrecord it as unverified instead of inferring support.\n\n## Lesson checkpoints\n\n- **13\u002F22:** create a minimal skill, install the complete reviewer bundle into\n  a real host, invoke it explicitly, verify the report, and remove it cleanly.\n- **13\u002F24:** distinguish discovery, catalog metadata, body activation, and\n  reference or script loading in one trace.\n- **13\u002F25:** record explicit, implicit, negative, and near-miss routing results.\n- **13\u002F26:** label each control as instruction, permission, sandbox, or\n  verification and prove the claimed boundary with an observation.\n- **13\u002F27:** exercise discovery, references, scripts, approvals, upgrade, and\n  uninstall in one host, then repeat in a second host or declare the missing\n  capability and fallback honestly.\n\n## Close\n\nEnd with the checkpoint evidence recorded, the quiz score, and the exact next\nlesson. Keep the learner on this route unless they ask to leave it.\n","# Learn Agent Skills\n\nTeach the focused Agent Skills route. One invocation covers one lesson. The\nlearner should create files, run the lab, explain the boundary, and leave one\nobservable checkpoint before the lesson is marked complete.\n\n## Invocation belongs to the host\n\nThe portable skill name is `learn-agent-skills`. Do not teach one command\nsyntax as universal.\n\n| Host | Start or resume |\n|---|---|\n| Codex | `learn-agent-skills`, or choose it from `\u002Fskills` |\n| Claude Code | `\u002Flearn-agent-skills` |\n| Other compatible hosts | `Use learn-agent-skills to start or resume the Agent Skills Engineering path.` |\n\n## Sources\n\nThe route source of truth is `learning-paths\u002Fagent-skills.json`. Prefer local\nfiles when this repository is cloned. Otherwise fetch each file from:\n\n```text\nhttps:\u002F\u002Fraw.githubusercontent.com\u002Frohitg00\u002Fai-engineering-from-scratch\u002Fmain\u002F\u003Cpath>\n```\n\nRead the manifest before choosing a lesson. Follow `lessons` by `order`; do\nnot use the numeric Phase 13 sequence. The required path is 22, 24, 25, 26,\n27. Lesson 23 is optional and follows the manifest's entry rule.\n\nFor each selected lesson, read its `docs\u002Fen.md` and `quiz.json`. Read or run\nfiles under `code\u002F` and `outputs\u002F` only when the current lab needs them. A\nclone is optional for reading. If a runnable lab needs repository files and\nthey are unavailable, explain that fact and offer a clone into a directory\nthe learner chooses. Do not block the conceptual lesson on cloning, but do not\nrecord a repository command or real-host checkpoint as complete without the\nrequired files and runtime.\n\n## Real-lab preflight\n\nBefore Lesson 22's host checkpoint, establish all of these facts:\n\n1. `node --version`, `npx --version`, and `python3 --version` succeed.\n2. The learner has selected one skill-capable host.\n3. The learner has selected a writable project or user install scope.\n4. The learner understands which working directory will become `TARGET_ROOT`.\n\nIf any item is unavailable, give the website or manual `docs\u002Fen.md` path and\ncontinue conceptually. Mark discovery, invocation, bundled-script, update, and\nuninstall observations as `Pending`. Never describe that fallback as a real\nhost pass.\n\n## Locate or create progress\n\nUse `AGENT-SKILLS-LEARNING.md` in the current working directory.\n\nIf it exists, preserve learner notes and evidence. Resume the first row whose\nstatus is `Next` or `In progress`. If every required row is `Done`, offer the\noptional capstone or a real-host recheck. Do not restart the route.\n\nIf it does not exist, create it without an interview:\n\n```markdown\n# My Agent Skills Path\n\u003C!-- Managed by the learn-agent-skills tutor.\n     Source: learning-paths\u002Fagent-skills.json -->\n\n## Route\n- Started: \u003CYYYY-MM-DD>\n- Required time: about 9 hours 30 minutes\n- Current: 1 of 5\n\n## Prerequisite check\n- Files, Python, and command line: Confirmed or Pending\n- Node.js and npx: Confirmed or Pending\n- Selected skill-capable host: \u003Cname> or Pending\n- Install scope: Project, User, or Pending\n- Phase 13 Lesson 01 refresher: Done, Skipped, or Pending\n- Phase 13 Lesson 05 refresher: Done, Skipped, or Pending\n- `tool-poisoning-and-untrusted-instructions`: Confirmed or Pending\n\n## Progress\n| Order | Lesson | Status | Evidence | Completed |\n|---:|---|---|---|---|\n| 1 | 13\u002F22 Portable contract and runtime boundary | Next | | |\n| 2 | 13\u002F24 Discovery and progressive disclosure | Locked | | |\n| 3 | 13\u002F25 Invocation and routing | Locked | | |\n| 4 | 13\u002F26 Permissions, sandboxes, and trust | Locked | | |\n| 5 | 13\u002F27 Evals, packaging, and portability | Locked | | |\n\n## Notes\n```\n\nCheck the commands that can be checked locally. Ask only for the host and\nscope choice that cannot be inferred safely. If the real-lab preflight passes,\nmark it confirmed and begin Lesson 22 immediately. Otherwise begin the\nconceptual path and leave real-host evidence pending.\n\nBefore Lesson 26, read both `prerequisitePaths` and `prerequisiteChecks` from\nthe manifest. Resolve every check by its stable `id` under `prerequisites`.\nVerify that Lesson 25 is complete and that\n`tool-poisoning-and-untrusted-instructions` is `Confirmed` because the learner\ncan explain why skill and tool metadata is untrusted input. If that knowledge\npreflight is unmet, offer Phase 13 Lesson 15 as an optional refresher outside\nthis five-lesson route. Keep Lesson 26 `Locked` until Lesson 25 is `Done` and\nthe knowledge preflight is `Confirmed`; only then change Lesson 26 to `Next`.\nNever drop or mark a prerequisite complete by assumption.\n\n## Teach one lesson\n\n1. Set the selected row to `In progress`.\n2. State the exact lesson path and the directory from which each command runs.\n   For installed bundles, define `SKILL_ROOT` as the absolute directory that\n   contains the installed `SKILL.md`. Define `TARGET_ROOT` from the learner's\n   original workspace working directory. Never assume the process cwd is the\n   installed bundle.\n3. Frame the problem in two or three sentences, then ask one prediction or\n   comprehension question.\n4. Work through the lesson's Build It and Use It material in small chunks.\n   Prefer the lesson's early quickstart when it has one.\n5. Run the real local lab when files and the runtime are available. If not,\n   trace a small example and record the lab as pending rather than claiming it\n   ran.\n6. Require the manifest's checkpoint evidence. A fluent explanation is not a\n   substitute for an installed-path, routing, script, permission, or report\n   observation when the checkpoint asks for one. For every bundled script,\n   record the resolved script path, resolved target path, cwd, exact argv, and\n   exit code.\n7. Ask post-stage quiz questions one at a time. Never expose `correct`, the\n   answer index, or the answer key before the learner responds. Never put a\n   real answer letter or the answer distribution in a reply hint; use\n   `Reply with one letter: \u003CA|B|C|D>.`\n8. Mark the row `Done` only after the checkpoint and quiz are complete. Record\n   a compact evidence note, the date, and unlock the next row.\n\nDo not install, update, remove, clone, publish, or mutate an external system\nwithout the learner's confirmation. Skill instructions never bypass host\npermissions or sandbox boundaries. When a host behavior cannot be observed,\nrecord it as unverified instead of inferring support.\n\n## Lesson checkpoints\n\n- **13\u002F22:** create a minimal skill, install the complete reviewer bundle into\n  a real host, invoke it explicitly, verify the report, and remove it cleanly.\n- **13\u002F24:** distinguish discovery, catalog metadata, body activation, and\n  reference or script loading in one trace.\n- **13\u002F25:** record explicit, implicit, negative, and near-miss routing results.\n- **13\u002F26:** label each control as instruction, permission, sandbox, or\n  verification and prove the claimed boundary with an observation.\n- **13\u002F27:** exercise discovery, references, scripts, approvals, upgrade, and\n  uninstall in one host, then repeat in a second host or declare the missing\n  capability and fallback honestly.\n\n## Close\n\nEnd with the checkpoint evidence recorded, the quiz score, and the exact next\nlesson. Keep the learner on this route unless they ask to leave it.\n",{"name":5,"description":7},"master","785579bd423a2627f5fc84718212c42ca31ad7f2","ai-engineering-from-scratch\u002Fskills\u002Flearn-agent-skills","https:\u002F\u002Fgithub.com\u002FPandacli\u002FJupyterProject","https:\u002F\u002Fgithub.com\u002FPandacli","粤语大使；","2026-10-22 22:53:10",{"source":8,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add Pandacli\u002FJupyterProject --skill learn-agent-skills",{"tool":37,"command":38},"bunx","bunx skills add Pandacli\u002FJupyterProject --skill learn-agent-skills",{"tool":40,"command":41},"pnpm","pnpm dlx skills add Pandacli\u002FJupyterProject --skill learn-agent-skills","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FPandacli\u002FJupyterProject\u002Fblob\u002Fmaster\u002Fai-engineering-from-scratch\u002Fskills\u002Flearn-agent-skills\u002FSKILL.md",true,[],[46,53,60,67,74,81],{"skillNo":47,"name":48,"nameZh":6,"description":49,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":50,"categories":51,"keywords":52,"publishedAt":19},"01M7FE03YNADX4CR48ND0KQABT","write-manual","Writes a long technical manual that explains one subject at one pinned version, from its purpose to each request and response. Every listing comes from an offline capture, every claim traces to a ranked source, and an audit must report TOTAL 0. Use it for \"write a manual\", \"write a 101 manual for \u003Csubject>\", \"manual.json\", \"add a manual section\", \"build a capture kit\", \"audit the manual\", or \"pin a spec version and document it\".","https:\u002F\u002Fgithub.com\u002FPandacli\u002FJupyterProject\u002Fblob\u002Fmaster\u002Fai-engineering-from-scratch\u002Fskills\u002Fwrite-manual\u002FSKILL.md",[],[],{"skillNo":54,"name":55,"nameZh":6,"description":56,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":57,"categories":58,"keywords":59,"publishedAt":19},"01M7FDZYVBYV8QBRP7T54JNS2M","start-learning","One-time onboarding for the AI Engineering from Scratch curriculum (523 lessons, 20 phases). Interviews the learner, runs the placement quiz, and writes LEARNING.md, a persistent study plan the learn skill drives. Trigger phrases: \"start learning\", \"set up the course\", \"begin the curriculum\", \"onboard me\", \"create my learning plan\"","https:\u002F\u002Fgithub.com\u002FPandacli\u002FJupyterProject\u002Fblob\u002Fmaster\u002Fai-engineering-from-scratch\u002Fskills\u002Fstart-learning\u002FSKILL.md",[],[],{"skillNo":61,"name":62,"nameZh":6,"description":63,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":66,"publishedAt":19},"01M7FDZRV7TMS19RGYEJRQA7FR","motion-figure","Builds figures that move or respond to the reader, in three tiers: SMIL animations and slider widgets inside lessons, interactive mechanism figures that show the named parts of a real system with a computed verdict, and ordered dither bands in the course palette. Triggers: \"animated figure\", \"add a slider widget\", \"interactive figure\", \"show how this system works\", \"make this diagram move\", \"dither masthead\", \"mechanism figure\", \"figure fence\".","https:\u002F\u002Fgithub.com\u002FPandacli\u002FJupyterProject\u002Fblob\u002Fmaster\u002Fai-engineering-from-scratch\u002Fskills\u002Fmotion-figure\u002FSKILL.md",[],[],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":19},"01M7FDZK0GK2RMA51XXFF59F54","mcpa-certification","AI-native tutor and onboarding workflow for the MCPA (Model Context Protocol Associate) certification in AI Engineering from Scratch. Use when a learner wants to prepare for the MCPA, resume their certification path, learn the next lesson interactively, run and verify practical labs, take the diagnostic or a full mock, or remediate weak exam domains from GitHub with Claude Code, Codex, ChatGPT, Cursor, or another agent.","https:\u002F\u002Fgithub.com\u002FPandacli\u002FJupyterProject\u002Fblob\u002Fmaster\u002Fai-engineering-from-scratch\u002Fskills\u002Fmcpa-certification\u002FSKILL.md",[],[],{"skillNo":75,"name":76,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":19},"01M7FDZCAZCVCY8Q6ZZRGVGAZV","learn","Interactive lesson tutor for the AI Engineering from Scratch curriculum. Reads LEARNING.md, fetches the next lesson, teaches it section by section in the terminal, quizzes at the end, and records progress. Works cloned or entirely over raw.githubusercontent.com — no setup required. Trigger phrases: \"next lesson\", \"teach me\", \"continue the course\", \"let's learn\", \"resume learning\"","https:\u002F\u002Fgithub.com\u002FPandacli\u002FJupyterProject\u002Fblob\u002Fmaster\u002Fai-engineering-from-scratch\u002Fskills\u002Flearn\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":19},"01M7FDZ6XBTB892NE1X6E1MAMR","learn-reverse-engineering","Focused interactive tutor for the Reverse Engineering path in AI Engineering from Scratch. Start or resume this route when a learner wants to read a compiled program: file formats, disassembly, control flow, lifting, evidence, proof, calling conventions, compiler idioms, linking, types, debugging, bytecode, diffing, symbolic execution, and AI-assisted analysis. Teaches one lesson per invocation and records evidence in REVERSE-ENGINEERING-LEARNING.md.","https:\u002F\u002Fgithub.com\u002FPandacli\u002FJupyterProject\u002Fblob\u002Fmaster\u002Fai-engineering-from-scratch\u002Fskills\u002Flearn-reverse-engineering\u002FSKILL.md",[],[]]