[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7B1NJ3GXFNQ22T66SAS47BC":3,"skill-same:01M7B1NJ3GXFNQ22T66SAS47BC":70},{"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":37,"commitSha":38,"skillDirectory":39,"forksCount":14,"repositoryHtmlUrl":40,"repositoryHomepageUrl":6,"ownerHtmlUrl":41,"ownerBlogUrl":6,"ownerDescription":42,"updatedAt":43,"install":44,"resources":57},"01M7B1NJ3GXFNQ22T66SAS47BC","install-agent-skills",null,"Install external Agent Skills repos collision-free.","dbillion\u002Fhermes-skills-public","dbillion","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F49393855?v=4","01M7AVH8QPGCM8WKWS22TRT0T6","Oludayo Adeoye","01M7AVH8RD1JJWA0YS9CFEWAV9",0,"MIT","https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Fsoftware-development\u002Finstall-agent-skills\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: install-agent-skills\ndescription: \"Install external Agent Skills repos collision-free.\"\nversion: 1.0.0\nauthor: Hermes Agent\nlicense: MIT\nmetadata:\n  hermes:\n    tags: [skills, install, github, collision-detection, agent-skills]\n    related_skills: [hermes-agent-skill-authoring, find-skills]\n---\n\n# Installing External Agent Skills into Hermes\n\n## Overview\n\nAgent Skills (Anthropic open-standard `SKILL.md` repos — e.g. manim, iart-ai,\nClaude-plugin marketplaces) drop into `~\u002F.hermes\u002Fskills\u002F\u003Ccategory>\u002F\u003Cname>\u002F`.\nInstalling looks trivial (`git clone` + copy) but has one sharp edge: **two\nskill directories that resolve to different real paths but share the same\nfrontmatter `name:` break skill loading** (the loader refuses to guess and\nreturns \"Ambiguous skill name\"). This skill makes installs collision-safe and\nsymlink-aware so a future session starts already knowing the trap.\n\n## When to Use\n\n- User pastes one or more GitHub URLs and says \"install this\" \u002F \"add these skills\".\n- You are wiring an external skill repo (manim, iart-ai packs, claude-plugin\n  marketplaces) into the local library.\n- You need to refresh an already-installed skill from upstream.\n\n**Don't use for:** authoring a brand-new skill from scratch (see\n`hermes-agent-skill-authoring`) or discovering *which* skill to use (see\n`find-skills`). This skill is the *mechanics* of getting an external repo's\nskills onto disk safely.\n\n## Workflow (ordered, with completion criteria)\n\n1. **Clone shallow** into a temp dir — never clone into `~\u002F.hermes\u002Fskills`\n   directly.\n   ```bash\n   cd \u002Ftmp && rm -rf skill_install && mkdir skill_install && cd skill_install\n   git clone --depth 1 \u003Curl> repo && echo CLONED\n   ```\n   *Done when* `find repo -name SKILL.md` lists the skill folders.\n\n2. **Enumerate candidate skills.** A \"skill\" is any directory containing a\n   `SKILL.md`. Repos often nest them under `skills\u002F`.\n   *Done when* you have the list of candidate `\u003Cdir>\u002FSKILL.md` paths.\n\n3. **Read each candidate's `name:` frontmatter field** (first `name:` line in\n   the YAML header). This `name` — NOT the directory name — is the loader key.\n   *Done when* every candidate has a resolved `name`.\n\n4. **Build the installed name→realpath map** (resolve symlinks!). Run\n   `scripts\u002Fdetect_skill_collisions.py` from this skill. It walks\n   `~\u002F.hermes\u002Fskills`, follows symlinks via `os.path.realpath`, and reports any\n   `name` that points to more than one distinct real path.\n   *Done when* you have the collision report (or \"NO COLLISIONS\").\n\n5. **Decide refresh vs new for each candidate:**\n   - **Name already maps to an installed realpath** → it's an UPDATE. Copy the\n     fresh repo content onto the *canonical* directory (the realpath the\n     symlink points to), do NOT create a parallel real directory.\n     ```bash\n     cp -rT \"$CANDIDATE\u002F.\" \"$INSTALLED_REALPATH\u002F\"\n     ```\n   - **Name is new** → install under a category folder with a disambiguated dir\n     name (repo + skill) to avoid dir collisions:\n     ```bash\n     mkdir -p ~\u002F.hermes\u002Fskills\u002Fcreative\u002F\u003Crepo>-\u003Cskillname>\n     cp -rT \"$CANDIDATE\u002F.\" ~\u002F.hermes\u002Fskills\u002Fcreative\u002F\u003Crepo>-\u003Cskillname>\u002F\n     ```\n   *Done when* every candidate is either refreshed in place or installed once.*\n\n6. **Remove any duplicate you created.** If step 5 left a stray parallel copy\n   (e.g. you copied into `creative\u002F` before realizing the name was symlinked\n   elsewhere), `rm -rf` the duplicate so only the canonical copy remains.\n   *Done when* `detect_skill_collisions.py` returns \"NO COLLISIONS\".*\n\n7. **Verify loads.** For each installed\u002Fupdated skill, call\n   `skill_view(name=\u003Cname>)` (or the categorized path if ambiguous) and confirm\n   `readiness_status: available`.\n   *Done when* all targets load without \"Ambiguous\" errors.*\n\n## Common Pitfalls\n\n1. **Name collision breaks loading.** The loader keys on `name:`, not the\n   folder. Two real dirs sharing a `name` → \"Ambiguous skill name … Refusing to\n   guess.\" Always map `name`→realpath *before* copying. (Real case: the\n   `manimce-best-practices` repo skill collided with a prior `npx skills add`\n   copy; refreshing the canonical dir + deleting the parallel copy fixed it.)\n\n2. **`.agents\u002Fskills` is the canonical install root for `npx skills add`.**\n   `npx skills add …` writes the real skill to `~\u002F.agents\u002Fskills\u002F\u003Cname>\u002F` and\n   symlinks it into `~\u002F.hermes\u002Fskills\u002F\u003Cname>`. If you `find` that path and see\n   \"no files\" but `diff` still reads it, it's a **symlink** — `find` without\n   `-L` does not descend into symlinked dirs. Resolve with `readlink -f` \u002F\n   `os.path.realpath` *before* deciding where to write, and refresh the\n   canonical target rather than creating a second real directory.\n\n3. **`find` symlink confusion.** Diagnostic trap: `find dir -type f` reports\n   nothing for a symlinked skill dir, yet `diff`\u002F`cat` read it fine. Trust\n   `readlink -f`, not `find`, to locate the real content.\n\n4. **`execute_code` may be gated** in some sessions (cron-mode approval).\n   If it returns a BLOCKED error, run the collision mapper as terminal python\n   instead — same logic, no dependency on the code tool.\n\n5. **Refreshing wipes nothing important if you `cp -rT`.** `-rT` (no trailing\n   slash on source, `--target-directory`) mirrors the source dir's contents\n   onto the destination, overwriting like-named files and leaving extras. It is\n   non-destructive to files the source doesn't contain — safe for a refresh.\n\n## Verification Checklist\n\n- [ ] Cloned with `--depth 1` into `\u002Ftmp`, not into `~\u002F.hermes\u002Fskills`\n- [ ] Every candidate `name:` parsed from frontmatter (not from dir name)\n- [ ] `scripts\u002Fdetect_skill_collisions.py` run against installed library\n- [ ] Updates copied onto the **canonical realpath** (resolved symlink target)\n- [ ] New skills installed once, under a disambiguated dir name\n- [ ] Any parallel duplicate removed\n- [ ] `detect_skill_collisions.py` now reports \"NO COLLISIONS\"\n- [ ] Each target `skill_view(name=…)` returns `readiness_status: available`\n\n## Support files\n\n- `scripts\u002Fdetect_skill_collisions.py` — re-runnable probe: maps every\n  installed skill `name`→realpath and prints collisions. Add\n  `--candidates \u003Cdir> …` to pre-test cloned repo skills before copying.\n- `references\u002Fmanim-install-example.md` — the worked manim\u002Fiart\u002Fyusuke install\n  that drove this skill: the exact collision found and how it was resolved.\n","# Installing External Agent Skills into Hermes\n\n## Overview\n\nAgent Skills (Anthropic open-standard `SKILL.md` repos — e.g. manim, iart-ai,\nClaude-plugin marketplaces) drop into `~\u002F.hermes\u002Fskills\u002F\u003Ccategory>\u002F\u003Cname>\u002F`.\nInstalling looks trivial (`git clone` + copy) but has one sharp edge: **two\nskill directories that resolve to different real paths but share the same\nfrontmatter `name:` break skill loading** (the loader refuses to guess and\nreturns \"Ambiguous skill name\"). This skill makes installs collision-safe and\nsymlink-aware so a future session starts already knowing the trap.\n\n## When to Use\n\n- User pastes one or more GitHub URLs and says \"install this\" \u002F \"add these skills\".\n- You are wiring an external skill repo (manim, iart-ai packs, claude-plugin\n  marketplaces) into the local library.\n- You need to refresh an already-installed skill from upstream.\n\n**Don't use for:** authoring a brand-new skill from scratch (see\n`hermes-agent-skill-authoring`) or discovering *which* skill to use (see\n`find-skills`). This skill is the *mechanics* of getting an external repo's\nskills onto disk safely.\n\n## Workflow (ordered, with completion criteria)\n\n1. **Clone shallow** into a temp dir — never clone into `~\u002F.hermes\u002Fskills`\n   directly.\n   ```bash\n   cd \u002Ftmp && rm -rf skill_install && mkdir skill_install && cd skill_install\n   git clone --depth 1 \u003Curl> repo && echo CLONED\n   ```\n   *Done when* `find repo -name SKILL.md` lists the skill folders.\n\n2. **Enumerate candidate skills.** A \"skill\" is any directory containing a\n   `SKILL.md`. Repos often nest them under `skills\u002F`.\n   *Done when* you have the list of candidate `\u003Cdir>\u002FSKILL.md` paths.\n\n3. **Read each candidate's `name:` frontmatter field** (first `name:` line in\n   the YAML header). This `name` — NOT the directory name — is the loader key.\n   *Done when* every candidate has a resolved `name`.\n\n4. **Build the installed name→realpath map** (resolve symlinks!). Run\n   `scripts\u002Fdetect_skill_collisions.py` from this skill. It walks\n   `~\u002F.hermes\u002Fskills`, follows symlinks via `os.path.realpath`, and reports any\n   `name` that points to more than one distinct real path.\n   *Done when* you have the collision report (or \"NO COLLISIONS\").\n\n5. **Decide refresh vs new for each candidate:**\n   - **Name already maps to an installed realpath** → it's an UPDATE. Copy the\n     fresh repo content onto the *canonical* directory (the realpath the\n     symlink points to), do NOT create a parallel real directory.\n     ```bash\n     cp -rT \"$CANDIDATE\u002F.\" \"$INSTALLED_REALPATH\u002F\"\n     ```\n   - **Name is new** → install under a category folder with a disambiguated dir\n     name (repo + skill) to avoid dir collisions:\n     ```bash\n     mkdir -p ~\u002F.hermes\u002Fskills\u002Fcreative\u002F\u003Crepo>-\u003Cskillname>\n     cp -rT \"$CANDIDATE\u002F.\" ~\u002F.hermes\u002Fskills\u002Fcreative\u002F\u003Crepo>-\u003Cskillname>\u002F\n     ```\n   *Done when* every candidate is either refreshed in place or installed once.*\n\n6. **Remove any duplicate you created.** If step 5 left a stray parallel copy\n   (e.g. you copied into `creative\u002F` before realizing the name was symlinked\n   elsewhere), `rm -rf` the duplicate so only the canonical copy remains.\n   *Done when* `detect_skill_collisions.py` returns \"NO COLLISIONS\".*\n\n7. **Verify loads.** For each installed\u002Fupdated skill, call\n   `skill_view(name=\u003Cname>)` (or the categorized path if ambiguous) and confirm\n   `readiness_status: available`.\n   *Done when* all targets load without \"Ambiguous\" errors.*\n\n## Common Pitfalls\n\n1. **Name collision breaks loading.** The loader keys on `name:`, not the\n   folder. Two real dirs sharing a `name` → \"Ambiguous skill name … Refusing to\n   guess.\" Always map `name`→realpath *before* copying. (Real case: the\n   `manimce-best-practices` repo skill collided with a prior `npx skills add`\n   copy; refreshing the canonical dir + deleting the parallel copy fixed it.)\n\n2. **`.agents\u002Fskills` is the canonical install root for `npx skills add`.**\n   `npx skills add …` writes the real skill to `~\u002F.agents\u002Fskills\u002F\u003Cname>\u002F` and\n   symlinks it into `~\u002F.hermes\u002Fskills\u002F\u003Cname>`. If you `find` that path and see\n   \"no files\" but `diff` still reads it, it's a **symlink** — `find` without\n   `-L` does not descend into symlinked dirs. Resolve with `readlink -f` \u002F\n   `os.path.realpath` *before* deciding where to write, and refresh the\n   canonical target rather than creating a second real directory.\n\n3. **`find` symlink confusion.** Diagnostic trap: `find dir -type f` reports\n   nothing for a symlinked skill dir, yet `diff`\u002F`cat` read it fine. Trust\n   `readlink -f`, not `find`, to locate the real content.\n\n4. **`execute_code` may be gated** in some sessions (cron-mode approval).\n   If it returns a BLOCKED error, run the collision mapper as terminal python\n   instead — same logic, no dependency on the code tool.\n\n5. **Refreshing wipes nothing important if you `cp -rT`.** `-rT` (no trailing\n   slash on source, `--target-directory`) mirrors the source dir's contents\n   onto the destination, overwriting like-named files and leaving extras. It is\n   non-destructive to files the source doesn't contain — safe for a refresh.\n\n## Verification Checklist\n\n- [ ] Cloned with `--depth 1` into `\u002Ftmp`, not into `~\u002F.hermes\u002Fskills`\n- [ ] Every candidate `name:` parsed from frontmatter (not from dir name)\n- [ ] `scripts\u002Fdetect_skill_collisions.py` run against installed library\n- [ ] Updates copied onto the **canonical realpath** (resolved symlink target)\n- [ ] New skills installed once, under a disambiguated dir name\n- [ ] Any parallel duplicate removed\n- [ ] `detect_skill_collisions.py` now reports \"NO COLLISIONS\"\n- [ ] Each target `skill_view(name=…)` returns `readiness_status: available`\n\n## Support files\n\n- `scripts\u002Fdetect_skill_collisions.py` — re-runnable probe: maps every\n  installed skill `name`→realpath and prints collisions. Add\n  `--candidates \u003Cdir> …` to pre-test cloned repo skills before copying.\n- `references\u002Fmanim-install-example.md` — the worked manim\u002Fiart\u002Fyusuke install\n  that drove this skill: the exact collision found and how it was resolved.\n",{"name":5,"author":24,"license":15,"version":25,"metadata":26,"description":7},"Hermes Agent","1.0.0",{"hermes":27},{"tags":28,"related_skills":34},[29,30,31,32,33],"skills","install","github","collision-detection","agent-skills",[35,36],"hermes-agent-skill-authoring","find-skills","master","f341c46fed021656175d5c9ccc67a72126e50bfb","skills\u002Fsoftware-development\u002Finstall-agent-skills","https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public","https:\u002F\u002Fgithub.com\u002Fdbillion","I enjoy writing codes,teaching programming, solving problems, tasted automation in Python and never looked back😋😋😋💞","2026-10-14 03:31:55",{"source":8,"skillName":5,"commands":45,"prompt":55,"zipAvailable":56},[46,49,52],{"tool":47,"command":48},"npx","npx skills add dbillion\u002Fhermes-skills-public --skill install-agent-skills",{"tool":50,"command":51},"bunx","bunx skills add dbillion\u002Fhermes-skills-public --skill install-agent-skills",{"tool":53,"command":54},"pnpm","pnpm dlx skills add dbillion\u002Fhermes-skills-public --skill install-agent-skills","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Fsoftware-development\u002Finstall-agent-skills\u002FSKILL.md",true,[58,64],{"path":59,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":62,"sourceHtmlUrl":63},"references\u002Fmanim-install-example.md","REFERENCE","参考文档",2883,"https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Fsoftware-development\u002Finstall-agent-skills\u002Freferences\u002Fmanim-install-example.md",{"path":65,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":68,"sourceHtmlUrl":69},"scripts\u002Fdetect_skill_collisions.py","SCRIPT","脚本",3772,"https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Fsoftware-development\u002Finstall-agent-skills\u002Fscripts\u002Fdetect_skill_collisions.py",[71,78,85,94,101,108],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":19},"01M7B25T0V7RBKES86Y3RTXQRN","blocked-page-recovery","Use when a fetch fails: 403\u002F429, paywall, WAF, bot wall.","https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Fweb\u002Fblocked-page-recovery\u002FSKILL.md",[],[],{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"publishedAt":19},"01M7B25DSBBMZRHVD28MAX7N5X","gemini-vision-fallback","Vision fallback: Gemini REST API if vision_analyze blocked.","https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Fvision\u002Fgemini-vision-fallback\u002FSKILL.md",[],[],{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":89,"licenseFileUrl":90,"sourceHtmlUrl":91,"categories":92,"keywords":93,"publishedAt":19},"01M7B22XW9QXT2ZWWXVZDFWYZP","twitter-cli","Use twitter-cli for ALL Twitter\u002FX operations — reading tweets, posting, replying, quoting, liking, retweeting, following, searching, user lookups. Invoke whenever user requests any Twitter interaction.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Ftwitter-cli\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Ftwitter-cli\u002FSKILL.md",[],[],{"skillNo":95,"name":96,"nameZh":6,"description":97,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":98,"categories":99,"keywords":100,"publishedAt":19},"01M7B22QMWY1B4F7BC114RG4S6","prism-scan","Structural analysis through dynamically generated cognitive lenses. Generates the optimal analytical lens for the specific code\u002Fartifact, then executes it. Finds conservation laws, structural invariants, and concrete bugs that vanilla analysis misses. Use on any code file, system design, or text artifact.","https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Fts-superhermes-prism-scan\u002FSKILL.md",[],[],{"skillNo":102,"name":103,"nameZh":6,"description":104,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":105,"categories":106,"keywords":107,"publishedAt":19},"01M7B22JA5XCAP1ACT6V50WYR1","prism-reflect","Constraint transparency: analyzes an artifact structurally, then analyzes what its own analysis concealed. Produces a conservation law AND a constraint report showing what was maximized, what was sacrificed, and what to investigate next. The only AI skill that knows what it can't see.","https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Fts-superhermes-prism-reflect\u002FSKILL.md",[],[],{"skillNo":109,"name":110,"nameZh":6,"description":111,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":112,"categories":113,"keywords":114,"publishedAt":19},"01M7B22EB15M75M4X3C52YGDCC","prism-full","Full Prism: multi-pass structural analysis with mandatory adversarial self-correction. Designs custom analytical passes, executes them with chaining, then attacks its own findings before synthesizing. Use for maximum depth on important code or artifacts.","https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Fts-superhermes-prism-full\u002FSKILL.md",[],[]]